Doris Tsao

Profil AI Expert

Nationality: 
Chinese
AI Speciality: 
Neuro Science
Current occupation: 
Professor, California Institute of Technology
AI Commitment Rate(%): 
40.88'%'
Twitter: 
https://twitter.com/doristsao
TwitterID: 
@doristsao
Tweet Visibility Status: 
Public

Description: 
Doris Ying Tsao is a neuroscientist recognized for her work on fMRI. She is very involved in the Artificial Intelligence community and participates in major debates in the sector. In 2017, his lab “cracked the code” of how our brains recognize faces, identifying dimensions of features that cause selective neurons in the face in different facial areas of the cortex to respond to faces. Thus, the images of faces presented to monkeys could be reconstructed with precision from the activity of selective facial neurons.

Aknowledged by:

Not Available

Les derniers messages de l'Expert:

Tweet list: 

2025-01-10 04:48:31 @AnimaAnandkumar @Caltech Relieved to hear this!

2025-01-10 04:33:00 @ZernickaGoetz Thinking of you, glad to hear you and your home remain safe.

2024-12-21 13:54:20 @KordingLab I often think about this question, in the context of: can we ever discover sth without even some kind of idea beforehand of what we are looking for? I come down on the side that at least in terms of broad frameworks of thinking, the answer is no. For those who argue yes, what is… https://t.co/Ji4pgAUrBn

2024-12-18 15:33:58 @ItzhakFried Einstein had all the puzzle pieces he needed--Michaelson Morley + Maxwell, both showing invariance of speed of light to reference frame. A better analogy may be to Aristotle...smart guy, but didn't have all the puzzle pieces.

2024-12-17 21:14:42 @NicoleCRust @_TheTransmitter An even bigger ratio: the ratio between the number of all possible 512 x 512 RGB images = 256^(3x512x512) = 10^1,893,914 GB, and the size of Stable diffusion model that can generate any image you can imagine = 4.2 GB. I think the ratios are related. The actual world &

2024-12-17 14:46:54 100% agree with @DavidAMarkowitz 's vision. But I would be much more specific in the lesson learned from the fly connectome revolution: we need to generate a *dense axonal projectome of the primate brain* NOW. This will give us the key sine qua non puzzle pieces to figure out… https://t.co/gN36sMPe5C https://t.co/l6FqVXeWGD

2024-12-13 19:53:17 Thank you @NicoleCRust for giving such a beautiful, inspiring talk and leading the discussion afterwards that made us all think outside our comfort zone. Really looking forward to continuing this conversation! https://t.co/bnDcdUIrl2

2024-11-23 06:14:39 RT @CoryMillerMarmo: The BRAIN Initiative has been an engine of innovation but is now fighting for its survival. Here we discuss why BRAIN…

2024-11-19 06:50:11 @pdhsu Yes, absolutely, I don't think how we learn, and why we desire to learn, is a random phenomenon, and I agree that it needs to be approached scientifically. It is perhaps the most important problem there is, since it determines what the next generation will be and therefore… https://t.co/WxaDG4fmPO

2024-11-18 14:49:58 @pdhsu Wow. What a different worldview on how to get children to learn. I'm thinking "mortal fear, sense of beauty, impossible dreams" and you're thinking "gamifying" and "modern social." I would love to have a salon on this topic!

2024-11-18 13:49:09 @pdhsu The most important question is how to make a child want to learn. I'm pretty sure this isn't it: "To keep the student on the rails without having to sit down next to them all the time, the adult needs to set up an incentive structure. Even little things go a long way, like "if… https://t.co/wHh6kKvpch

2024-11-17 00:25:43 RT @DavidAMarkowitz: The $3B Human Genome Project was started under a Republican administration and later acknowledged by the Obama admin t…

2024-11-16 14:58:32 Does anyone have any recommendations for where I can find a really talented blender scripter? Please email me dortsao@berkeley.edu

2024-11-09 22:06:51 @NicoleCRust @seeingwithsound @SimonsFdn @yael_niv Astonished by your meta-morphosis &

2024-11-01 22:15:22 @DulacLab @MCB_Harvard What wonderful news! Congratulations, Catherine!

2024-10-24 14:03:03 RT @Pieters_Tweet: This new paper spearheaded by Matt Self shows the link between border-ownership tuning in area V4 of the visual cortex a…

2024-10-22 12:56:02 RT @NINDSdirector: NIMH Director search is now open. Country is looking for someone totally devoted to focusing NIH resources on the resea…

2024-10-20 16:03:30 @LilaDavachi @gershbrain @martisamuser @rungtalab Fully agree that fmri gives incredibly valuable whole brain insights!!! None of our work on face patches would have been possible without fmri. Not sure what you mean by "single unit approach had worn out its utility." I think there may be confusion about what's meant by "single… https://t.co/N5WiphPAH6

2024-10-20 15:27:07 @LilaDavachi @gershbrain @martisamuser @rungtalab I would argue that all of these phenomena ultimately arise from firing of many single neurons, so ultimately we need to record from many single neurons. Even if behavior X correlated perfectly with fmri activity in area Y and the latter was furthermore necessary for behavior X,… https://t.co/CEgqmUNTwh

2024-10-20 14:09:50 @LilaDavachi @gershbrain @martisamuser @rungtalab Hmmm? Are you saying the jury is still out on whether perception and behavior arises from electrical activity of neurons?

2024-10-19 14:06:56 @gershbrain @VincentCostaPhd Does this issue really exist? I haven't noticed it. The systems neuroscientists I talk to who care about perception and cognition--which obviously can't be understood by only studying neurons in a single mm of cortex--are aware of and deeply interested in human neuroimaging… https://t.co/G18xNuN6lU

2024-10-18 17:15:08 @gershbrain @martisamuser Hmmm...but the relationship could be *extremely* indirect. Had an interesting discussion with @rungtalab recently, who mentioned that neuromodulators like acetylcholine and noradrenaline can have direct effects on blood vessels. So the ultimate neural cause of a BOLD signal in V1… https://t.co/cAxAervDsQ

2024-10-18 00:02:24 RT @UeliRutishauser: Super excited about our new result. Thanks to both @WadiaVarun and @doristsao for making this incredible collaboration…

2024-10-17 01:20:18 Super excited about @WadiaVarun ’s new preprint providing what I think is the most compelling single-neuron evidence for the existence of a generative model in the human brain. Varun first showed that human VTC uses the exact same single-cell coding principles as macaque IT… https://t.co/xLVtP97MXK https://t.co/dEOoHOrWPV

2024-10-16 02:20:28 RT @nanoassembly: Deadline Nov 1 https://t.co/RI8JOOkll1 https://t.co/fzOtuRlmoS

2024-10-08 15:07:25 @SebastianSeung Wow, I didn't know that! I loved Longuet-Higgins' book 'Mental Processes,' I can see how he must have been an inspiring advisor.

2024-10-08 14:14:52 https://t.co/FvTacysIjo

2024-10-08 13:15:17 RT @Yoshua_Bengio: @HopfieldJohn and @geoffreyhinton, along with collaborators, have created a beautiful and insightful bridge between phys…

2024-10-08 12:04:31 Omg!!! I did a double take when I saw 'Nobel Prize,' and figured @NicoleCRust was making a very cool joke, because John Hopfield certainly deserves the Nobel Prize. Then I heard the news. What a right and perfect and monumentally inspired choice, celebrating the boundary… https://t.co/jy1HCEOgwM https://t.co/BavTTWVmzk

2024-10-07 11:23:00 This (Tuesday) afternoon at #sfn2024 ! Come check out @HesseJanis talk about his work on conscious perception in MCP Room S106!4:00 PM - 4:15 PM. NANO32.13 - Dynamics of representations of conscious perception and physical stimulus across inferotemporal, parietal and prefrontal… https://t.co/5RKbe8h4pE

2024-09-19 15:18:10 So valuable! Embarrassing how many times we have had to"rediscover" this https://t.co/3rC2zTMvoG

2024-09-17 22:06:31 @neuralink Congratulations Dan and team!!! This is incredible news.

2024-09-13 17:31:40 RT @tylerraye: do large-scale vision models represent the 3D structure of objects? excited to share our benchmark: multiview object consis…

2024-09-11 04:51:29 RT @BrainsExplained: Remember how many of us were super upset about NIH's proposal to limit K99 eligibility to 2 years? If you never manag…

2024-09-11 01:04:07 @djfreedman David! Wow, congrats!! Very exciting for UChicago

2024-09-10 13:35:20 @ElShamayleh Yes, absolutely!

2024-09-10 13:11:02 Great job opportunity to start a neuroscience lab at the best public university in the world https://t.co/yfurEnOkGF

2024-09-07 21:44:19 This past week in Norway for Kavli Prize Week felt like a lucid dream—a transformative experience of cultural discovery, scientific camaraderie, and moments of deep personal reflection. From arriving in Oslo and being whisked away to the historic Grand Hotel, to the awe-inspiring… https://t.co/cNg2uVspSt https://t.co/w6aZkC1TKA

2024-08-22 18:34:52 @SergeyStavisky Thank you, Sergey! I suspect in the not far future the worlds of vision &

2024-08-21 00:25:23 RT @patrickmineault: I'll be speaking at the NIH neuroethics working group this Wednesday, 12:30 EST on foundation models for neuroscience,…

2024-08-15 14:42:02 @madsarv @ardemp @WiringTheBrain @NicoleCRust @seanmcarroll @alvanoe Wonderful! Yes, I am super excited he will be coming to the meeting!

2024-08-15 14:09:18 RT @doristsao: @ardemp @WiringTheBrain @NicoleCRust @seanmcarroll Yes! One of my favorite books about consciousness, "Action in Perception"…

2024-08-15 14:07:27 @ardemp @WiringTheBrain @NicoleCRust @seanmcarroll Yes! One of my favorite books about consciousness, "Action in Perception" by @alvanoe, argues that the process of piecing together the shape of an object from a sequence of touches, via a *skillful sensorimotor loop* is the essence of consciousness. "On the enactive view, all… https://t.co/Moz7eqG1xu https://t.co/wL0CV4CWlP

2024-08-15 13:36:39 @DrDavidBrandman @NS_Card Congratulations, incredible!

2024-08-15 13:13:13 @NicoleCRust @jmacshine I think of voluntary control of behavior and in particular, the preSMA, sandwiched between prefrontal cortex and M1. Stimulation of preSMA produces an urge to move, and lesions can cause uncontrolled actions. https://t.co/chUeRCZpGB https://t.co/6OIhfpSAIg

2024-08-14 05:04:30 @NicoleCRust Thank you, this means so much coming from you! I do believe all of us studying the brain are contributing to understanding consciousness. The feedback circuits that you highlighted in your talk (which is still buzzing in my head...so lucid &

2024-08-08 02:52:33 Wow. This was awe-inspiring--both on a personal level, seeing an example of a scientist refusing to simply respond to the day-to-day &

2024-08-07 01:00:05 Can't wait to see what this amazing team develops! https://t.co/XEP3AGolxE

2024-08-06 20:57:57 What a beautiful, truly visionary talk by @MatthiasBethge! Resonate so much with the idea that knowing how to parse incoming information so that we update our world model appropriately (#1-3 below) is a key missing piece in our current understanding of the brain.… https://t.co/HT3MmA2d1l https://t.co/pFekVhlIpz

2024-08-05 14:05:29 An exciting opportunity https://t.co/LV2bdbePMw

2024-08-04 14:19:32 @BlumLenore Thank you Lenore! I wish I could join your walk in person

2024-07-31 16:52:04 RT @ardemp: I’ve been lately listening to podcasts during my early morning runs. Today was so good, I ended up walking just to finish it. T…

2024-07-29 23:12:05 https://t.co/t53pgnD4NP

2024-07-29 13:01:44 It was an honor and joy to chat with Sean Carroll. We talked about vision, consciousness, the bridge between sensory and symbolic processing, and a few other little things in this preposterous universe... https://t.co/YGERThBMf2

2024-07-28 02:43:37 RT @MichaelLinLab: There's already too much tunnel vision in our funding systems, and forcing postdocs to apply to K99s within their first…

2024-07-23 14:04:09 Congratulations to the incredible new group of HHMI Investigators! Take us to the stars! https://t.co/7k4jzjvVLE

2024-07-17 00:56:59 RT @thefrankslab: Please RT. The Franks Lab at Duke U. is hiring a neuroscience tech! We study perception, learning &

2024-07-16 16:14:39 RT @DrAnneCarpenter: US citizens or permanent residents from an under-represented group! You can contact any NIH-funded researcher you adm…

2024-07-13 19:06:54 @smfleming Strongly agree that theory needs to be pursued hand-in-glove with experiment. Consciousness science needs its version of black body radiation, 2-slit experiment, etc--bona fide empirical mysteries concerning the concrete workings of conscious systems. IMHO, the consistency of… https://t.co/b8QX8oH8jM

2024-07-13 16:23:08 @ma_fakharian @reziliusReza @JDoostmohammadi Thank you both for clarifying!

2024-07-12 16:04:47 @gershbrain Love this question. This also bothers me a lot and I would love to read a lucid treatment. To me, it's related to the question of whether there is a unique, correct computational interpretation of a physical system, going beyond simplicity. I think the fact that consciousness… https://t.co/FoqbLmgdbR

2024-07-11 01:27:29 As our country debates the capabilities of a particular neural network, the importance of fundamental research in neuroscience is clearer than ever. https://t.co/0apwcjngrK

2024-07-08 18:56:39 Incredibly impressive and exciting development for NHP neuroscience! https://t.co/99ILbFmuKy

2024-07-07 01:44:59 @BlumLenore @cstheory @MIT What an extraordinary picture (whose meaning I could only begin to understand after hearing your and Manuel's recent talks at Simons).

2024-07-05 22:29:55 We are searching for a new lab manager. Great opportunity to learn about visual neuroscience, NHP research, and cutting-edge techniques in electrophysiology, fMRI, and data analysis. Excellent springboard to grad/med school. Email me if you are interested.

2024-07-02 14:04:40 If you are working in neuroAI (or simply curious about it) you should come. This meeting is going to be amazing, check out the speakers https://t.co/ZGF6Gczi6s

2024-06-26 17:49:53 Congratulations to my amazing, endlessly creative, and fearless colleague Yang Dan! https://t.co/czRSJ38pQh

2024-06-25 12:17:13 RT @neurovium: The neuron as a direct data-driven controller Mitya's @chklovskii paper out in @PNASNews https://t.co/jMczN8xsBN

2024-06-24 17:21:06 @YiMaTweets Your energy and care for teaching young people is so inspiring! Wish you can come back to Berkeley!

2024-06-22 14:06:06 RT @doristsao: @ylecun @Meta Very interesting! One argument I find compelling for why the brain *does* have a generative model that fills i…

2024-03-01 00:00:00 CAFIAC FIX

2024-03-11 00:00:00 CAFIAC FIX

2023-05-19 19:00:00 CAFIAC FIX

2023-05-21 19:00:00 CAFIAC FIX

2023-04-21 00:00:01 CAFIAC FIX

2023-04-07 04:07:08 @lisa_giocomo Congratulations, Lisa!!!

2023-04-04 03:12:22 @TheBrunoCortex @Ryohei_Neuro @mattlark @FritjofHelmchen

2023-03-24 17:51:57 RT @NatureNeuro: A remembrance of Krishna Shenoy from @MarkChurchland and Paul Nuyujukian https://t.co/fYnlWci1PM

2023-03-24 17:49:32 @NatureNeuro @MarkChurchland This was so beautiful and thought provoking. The challenge of dealing with your own ego is probably one of the hardest aspects of being a scientist, and it is so incredibly inspiring how Krishna took a complete left turn where everyone else goes right.

2023-03-23 19:18:06 @pfau or, you're secretly afraid that intelligence is not that complicated

2023-03-23 19:05:39 @smfleming Thank you! I'm so glad someone has read it...

2023-03-23 19:02:34 @drkjjeffery @WiringTheBrain @FENSorg as it implies a mental 'I' that is distinct from both the brain and the body but that can nevertheless trigger brain events and, thus, bodily movement."

2023-03-23 19:02:21 @drkjjeffery @WiringTheBrain @FENSorg From Haggard NRN 2008: "A dualistic view of endogenous causation is engrained in our normal language. Everyday language suggests that 'I' consciously choose to perform actions and that 'my' choice somehow causes the action to occur. This language is dualistic, ...

2023-03-23 18:47:27 @WiringTheBrain @FENSorg Was revelation to me that the very use of the pronoun "I" betrays a naive dualist view.

2023-03-23 18:47:00 @WiringTheBrain @FENSorg Why not the mind body problem, which would cover all of these topics? In teaching neuroscience, philosophy &

2023-03-20 16:29:53 Inspiring story and a model we clearly should try to rekindle here at Berkeley. Love the emphasis on preparing students for high achievement and not just passing. https://t.co/vckPQVzwKo

2023-03-14 17:05:01 @AndrewHires @USC Yay Andrew!!!

2023-03-09 23:13:36 RT @SuryaGanguli: A remarkable ability of #ChatGPT to solve a simple geometric reasoning problem, reveal hidden assumptions when asked, and…

2023-03-09 08:06:22 @platobooktour The point that humans learn much more efficiently is important and inarguable. But I think what we all wanted to know was, "Isn't Chomsky amazed by LLMs?" And his answer suggesting he hadn't even played with them was heartbreaking...in a child-realizing- dad-is-only-human way.

2023-03-07 18:48:11 @GaryMarcus If true conscious bladerunner-level AGI emerges, I think having our jobs taken away doesn't rank in the top 5 of problems we will need to face

2023-03-05 10:00:00 CAFIAC FIX

2023-03-02 22:00:00 CAFIAC FIX

2023-02-27 01:00:00 CAFIAC FIX

2023-02-10 16:46:39 @MelMitchell1 To be clear: you mean no one has figured out if dalle, chatgpt etc have figured this out? Or no one has figured out how to do this, period?

2023-02-10 16:38:49 @MelMitchell1 The part I liked best was when he explained that best way to compress a list of additions is to figure out the algorithm for addition and write code to do that. Rules of addition for generative models = compositional structure of images, language. Q: have they figured this out?

2023-02-08 23:23:10 @Plinz Are you building one?! Can I join? ...I don't know about name, but the front gate should be inscribed with Roy's monologue

2023-02-08 17:41:39 @NeuroPolarbear @BCMNeurosurgery @SameerShethMD @KBijanki @AndrewJWatrous @SRHeilbronner How exciting (for you, for Baylor, and for human neuroscience)! Congratulations!

2023-02-08 16:38:53 @reziliusReza @ma_fakharian @just_alden Beautiful! Was the cell actually tuned to the direction of movement, or did it simply respond every time there was an eye movement?

2023-02-08 15:49:22 @Nancy_Kanwisher It's interesting that you put V1 in the non specialized category but the various IT regions in the specialized category. To me, they are actually very similar, with V1 representing a map of retinitopic space and IT a map of object space.

2023-02-07 16:28:09 RT @xsteenbrugge: "Voyage through Time" is my first artpiece using #stablediffusion and I am blown away with the possibilities... We're cr…

2023-02-06 22:13:13 @HesseJanis and Frank Lanfranchi in our lab, together with Daniel Wagenaar at Caltech, moved mountains to figure out a reliable way to insert these probes deep into the temporal lobe (e.g., face patch AM) without breaking them. https://t.co/648FoimvcE

2023-02-06 22:02:26 These hierarchical maps are the basis for decomposing the structure of the world. But I agree 100% there is a huge piece missing in our understanding, of how the elements are then recomposed, to build a complete compositional understanding of the world--a world model.

2023-02-06 22:01:47 Re the last: evolution figured out to have 5 senses, that capture independent aspects of the world (taste and smell sense chemical structure, vision senses spatial organization of matter). And deeper in the brain, evolution figured out how to build maps and maps atop maps.

2023-02-06 22:01:05 I once heard @ylecun give a talk where he said 'either god exists or the world is compositional.' I love this so much--it explains everything from the reductionist basis for science to the reason why generative models of reality can exist to why the brain is modular. https://t.co/Pi14K5xRr4

2023-02-06 21:18:09 Very grateful to have been part of this collaboration, and happy this is out! My lab essentially doesn't do single tungsten electrode recordings anymore, except for initially verifying targeting. The Neuropixel 1.0-NHP probe really is a game changer https://t.co/NgU4hhxYNg

2023-02-06 15:27:24 @PhilCorlett1 Come hang out with us at Berkeley! https://t.co/Z4PVIPg8Nt

2023-02-02 04:25:36 @PhilCorlett1 Congratulations Phil!

2023-01-30 01:00:00 CAFIAC FIX

2023-01-12 23:52:52 RT @Li_Zhaoping: our brand new summer school on "Systems Vision Science", see https://t.co/KbVq3I1Ogc please RT, https://t.co/jTg90PbkGp

2022-12-31 23:27:59 Are there any cool New Year's Eve parties happening tonight in the Bay Area? Asking for a friend

2022-12-21 17:29:17 @davidchalmers42 spaceship

2022-12-20 19:32:58 RT @TonyZador: Here is the obituary i wrote about my postdoc advisor Chuck Stevens who died in October at the age of 88. He was a brillia…

2022-12-17 01:13:52 A beautiful day at Cambridge celebrating the intellectual legacy of Horace Barlow, starting with a sublime tribute from @mameister4. He showed # citations of 1961 paper = ReLU(year-1961-30). That is the rarest kind of scientific greatness.

2022-12-09 19:01:00 @Chan_Rec_Transp @emily_liman Congratulations Emily!!!

2022-12-08 14:50:16 @pfau Which is the more "powerful" part of a computer, the power supply or the processor?

2022-12-08 13:00:00 CAFIAC FIX

2022-12-07 08:00:00 CAFIAC FIX

2022-11-04 15:20:50 @ScienceMagazine @Antihebbiann Yay!!! Congratulations Ann!

2022-11-01 18:18:13 @hans_opdebeeck Indeed! What remains an open question for me is whether the axis for coding specific face identity is the same as that discoverable with responses to nonface objects, or whether there is specialization for coding face identity in face patches.

2022-11-01 16:55:29 @hans_opdebeeck PS Here is one key figure from Bao et al. (ED 10) showing that object space projection better explains activity than semantic label as body. Group 2 and Group 3 objects had similar projection onto the "body" quadrant, and they also evoked similar fmri &

2022-11-01 16:32:51 @hans_opdebeeck The reason Hubel &

2022-11-01 16:25:02 @hans_opdebeeck We called the axes 'stubby-spiky' and 'animate-inanimate' because it seemed useful to have a term besides PC1 and PC2

2022-11-01 16:22:27 @hans_opdebeeck What we did propose is that it is organized according two first two axes of object space that can be discovered by a DNN, and that responses of cells in the macaque body patch are better explained by projection onto the 'body' quadrant of this space than by category label

2022-11-01 16:21:22 @hans_opdebeeck Hi Hans, looking forward to reading your paper in detail! I just want to clarify that I have never proposed that IT cortex is organized according to animacy x aspect ratio.

2022-10-31 20:28:21 RT @dileeplearning: it is perception that needs the breakthrough, not the other things. If perception was solved all other things would bec…

2022-10-29 14:57:06 @gordonneuro @ndosenbach @DrDamienFair @GrattonCaterina @DeannaJGreene @smarek0502 @RoselyneChauvin @BarchDbarch @ChadMSylvester @DrAshRajesh @dillannewbold Very interesting, congratulations! Underscores the dire need for full human and macaque projectomes--will unleash so many discoveries.

2022-10-27 15:53:48 RT @Nancy_Kanwisher: Marge's work is hugely important scientifically and clinically, answering fundamental questions about brain developmen…

2022-10-27 15:50:39 I have seen how deeply and personally Marge cares for the animals in her lab, she is truly the last person who should be targeted. https://t.co/LgivwWDFTE

2022-10-26 19:02:31 On cloud 9 after reading this beautiful, brilliant perspective by Qasim Zaidi on 3d vision. An homage to Poincare’s insight that the origin of 3d vision is the isomorphism between the groupoid of transformations between perspectives &

2022-10-25 19:14:21 @tyrell_turing @NSERC_CRSNG Congrats Blake!!!

2022-10-21 23:17:53 https://t.co/DbWFqFnPR4 https://t.co/5597bH5ojO

2022-10-11 16:35:58 Very excited about this symposium in Cambridge to celebrate Horace Barlow https://t.co/MoGOotKCY3

2022-10-06 19:53:29 I am so happy to finally see this in print: https://t.co/DN9C1Nbvx1 Next step: to see this in the brain https://t.co/tKzPKkdDSB

2022-10-03 19:44:00 @AdrienDoerig Thank you for clarifying!

2022-09-27 11:44:37 @AdrienDoerig Maybe I should phrase it differently: what is your prediction wrt coding of in single cells in IT, in, say a face patch and a spiky patch?

2022-09-27 11:36:42 @AdrienDoerig good model of the general rule in VTC. And I thought your paper is arguing that the representation in VTC is not of individual objects but also of context in which objects are placed.

2022-09-27 11:34:38 @AdrienDoerig I would think we are constantly segmenting any object that we are looking at, as well as nearby ones, and processing each individually. At the same time, we are also processing semantic relationships between them. So it seems what happens in a face patch wrt context should be

2022-09-27 05:17:29 @AdrienDoerig I was thinking of the former...add a person (with a frontal face in fovea) to scenes from your database without any people. Is it your prediction that face cells should care a lot about background content, and moreover variance should be explained by a semantic model?

2022-09-27 04:56:25 @AdrienDoerig Very interesting! So, if we took all your images without faces, and inserted the same person with a face in them, would you predict that we should explain a lot of variance in face cell responses using your semantic embedding? I am somehow skeptical but would be amazed if true

2022-09-24 03:05:30 RT @nanoassembly: #Reviewer2

2022-09-21 13:49:32 @flodlan @m_heilb @YiMaTweets *need not be

2022-09-21 13:45:16 @flodlan @m_heilb @YiMaTweets Btw, Yi's argument (https://t.co/X2kEem5ftB) is that if the forward path f is expansive then you can detect difference at the end. And measuring difference at the end makes sense because the difference will be meaningful ... eg tiny shift in position will be ignored

2022-09-21 13:30:48 @flodlan @m_heilb Fully agree that error need to be measured at each stage by explicit prediction error neurons (and indeed it may make more sense to measure at the end, after internal model is built, as @YiMaTweets argues) But some kind of error needs to be measured, at some stage, no?

2022-09-21 13:27:22 @flodlan @m_heilb "Also, researchers studying predictive processing often assume the existence of two types of neurons, prediction and prediction error neurons...However, predictive processing models without explicit error representation have also been defined"

2022-09-14 15:38:41 RT @NatRevNeurosci: The neural bases for timing of durations – a Review by Albert Tsao, S. Aryana Yousefzadeh, Warren Meck, May-Britt Moser…

2022-09-14 11:17:48 @davidchalmers42 Possibly none. @YiMaTweets explains, in a GUT of ML, how just two (very old) ideas, parsimony and self consistency, can explain almost all the networks du jour mentioned here, including transformers, contrastive learning, GANs... https://t.co/X2kEem5ftB

2022-09-09 05:25:01 Thank you @UeliRutishauser for taking us along your journey to understand memory, your talk was truly extraordinary. It was great to hang out again! https://t.co/9rGduZEHs5

2022-08-28 15:48:20 @mikhailshapiro Had to search for more than a minute through the crowd, but so good to see that smile again! :)

2022-08-28 14:49:02 RT @davidchalmers42: the #ASSC26 keynote speakers are super+:YOSHUA BENGIO (deep learning/ consciousness)JOE LEDOUX (@theamygdaloid, fe…

2022-08-18 19:58:47 RT @dyamins: 1/ Excited to announce EISEN, our new work on self-supervised, category-agnostic instance segmentation: EISEN learns to segmen…

2022-08-07 01:38:47 @2ittingMammad @WiringTheBrain Planning in progress, very grateful for this thread!

2022-08-06 17:27:42 @WiringTheBrain Wow, thank you! Some of my favorites: Land of Silence and Darkness (the space of brains), Rashomon (the nature of certainty), Florence Foster Jenkins (free will versus genetic determination)

2022-08-06 15:32:02 @WiringTheBrain I'm teaching a new class in the spring on neuroscience, film, and philosophy, where we watch movies, talk about brains, and speculate about philosophical conundrums. Thank you for writing this great course syllabus! :)

2022-07-22 08:02:50 RT @neuroamyo: I've noticed my tweets about my dog often get more traction than those about my science. So, here's my dog. Also, we're hiri…

2022-07-16 15:51:45 @MelMitchell1 @AlisonGopnik When Helen Keller described her epiphany as 'I suddenly saw everything with the strange, new sight that had come to me,' was she simply completing narratives generated by sighted people a la lambda, or did she suddenly gain power to control her internal generative model?

2022-07-16 00:02:07 @_Chen_Ran_ @SCglobalbrain Congratulations Chen!

2022-07-15 23:10:32 Congratulations to the new fellows!!! Am I allowed to brag that I was turned down by *two* of this distinguished group for postdocs? :) https://t.co/iFAa5LylQb

2022-07-15 21:41:11 @JeremiahYCohen Congrats Jeremiah, glad to have you on west coast!

2022-07-15 15:11:16 @dacolon @Neuro_CF How wonderful! I had similar experience years ago in Shanghai, was rummaging in violin shops and stumbled into ION. I ventured in and found myself in a lab with a printout of David Hubel taped to the door.

2022-07-14 20:44:53 @AdamHantman Interesting! I would guess there is a loop between proprioceptive signals and internal state that is purely perceptual, and it's not all RL, and so the insights would generalize. What the structured internal representation should be, I don't know

2022-07-12 15:54:42 @MolemanPeter @YiMaTweets @harryshum Indeed! There may be more than one loop (e.g., dorsal &

2022-07-12 05:13:44 I love systems of thought and works of art that feel inevitable. (17/n)

2022-07-12 05:13:16 And I haven’t even mentioned the engineering feats that Yi and his students have already accomplished with this new scheme, including solving the problem of catastrophic forgetting https://t.co/7mUTzJ7aI3 (16/n)

2022-07-12 05:12:55 Moreover, it points to many exciting new directions, particularly in applying this approach to achieve 3D vision and intuitive physics, where the low-dimensional constraints are likely stronger than the soft statistical correlations among static images. (15/n)

2022-07-12 05:12:36 What’s beautiful is how it weaves all these puzzle pieces together starting from first principles and shows how everything can be made computationally tractable (Yi calls the measurement of rate reduction “bean counting”). It’s an amazing jigsaw puzzle. (14/n)

2022-07-12 05:12:19 The framework has resonances to many previous ideas including sparse coding, redundancy reduction, disentangling, predictive coding, generative models, feedback, Bayesian inference, free energy, wake-sleep algorithm, contrastive learning, transformers… (13/n)

2022-07-12 05:11:46 It also implies (harder to see) that the interaction between f and g should be through a pursuit-evasion game in which they play opposite sides of a joint objective function, rather than an autoencoder (for a simple linear algebra reason explained in paper). (12/n)

2022-07-12 05:11:25 If true, this implies that visual information should make three passes through each area during learning! (11/n)

2022-07-12 05:11:05 The paper argues that in this loop, error should be measured in the internal space, between z = f(x) and ẑ= f(g(f(x))). After all, the whole job of the feedforward path f is to build a meaningful measure of difference. (10/n)

2022-07-12 05:10:36 Thus we arrive at the following general architecture for perception/intelligence: a closed loop between a compressor f and a generator g. (9/n) https://t.co/MP90E6Q9Sd

2022-07-12 05:09:58 How can we ensure that our compressed representation completely and faithfully represents the original data? For this we need a second principle, the Principle of Self-Consistency, motivated by the general notion of error correction through feedback. (8/n) https://t.co/cqXScbHAg0

2022-07-12 05:09:36 Surprising result: this general definition of parsimony through rate reduction precisely specifies the optimal internal representation. Indeed, one can build a “white box” feedforward network that unrolls to maximize rate reduction w/o needing any back prop! (7/n) https://t.co/EPm9Ibhut4

2022-07-12 05:08:15 This naturally suggests a simple geometric formula for measuring parsimony, rate reduction. The picture below shows data embedded in two linear subspaces S1, S2 within the latent space. Rate reduction = log (ratio of the # (blue + green) to # green spheres). (6/n) https://t.co/nyTiD2HI4g

2022-07-12 05:07:51 But how can we achieve a parsimonious representation? Should it be the one with minimum description length (as in Hinton 1995)? No! We should choose a measure that captures the form of parsimony we care about, structured low-dimensional representations (cf Brette 2019). (5/n)

2022-07-12 05:07:25 The paper begins by asking, what is the basis for intelligence? Answer: Low-dimensional structure in the environment. This is what enables prediction and generalization. Ergo the Principle of Parsimony (4/n) https://t.co/pipzrhDpII

2022-07-12 05:06:55 I think one must read the paper to grasp its impact. Here are some of the highlights of the paper for me. (3/n)

2022-07-12 05:06:39 Indeed, it sheds light on almost every question I’ve ever been interested in. Yi asked me for comments. This resulted in multiple all-day conversations where he patiently explained each part that was unclear to me. For this, I was rewarded with a co-authorship! (2/n)

2022-07-12 05:05:56 A month ago, an amazing gift fell in my lap. @YiMaTweets sent me his manuscript with @harryshum on the two principles underlying Intelligence. This is a monumental paper, clarifying the puzzle of how intelligence arises through a beautiful series of deductions. (1/n) https://t.co/oRs2PIOICx

2022-06-19 22:36:04 RT @nanoassembly: As an undergrad I wanted to work on my ideas but had to instead work in potato fields to help my single mom raise me and…

2022-06-15 00:01:18 @PhilCorlett1 @stevewcchang His remarks were sublime. I remember a magical feeling of communion among all of us after he spoke...am so grateful for that dinner, though the memory will now always be tinged with sadness

2022-06-10 18:18:28 @MrJeffKarp Jeff, sorry to hear you're going through this, take care and hope you recover fully soon!

2022-06-10 03:29:36 I am heartbroken to hear about Gordon Shepherd's passing. He just emailed me two days ago with ideas about the commonalities between olfaction and vision, and sent a picture from my recent visit to Yale. A towering scientist and human being. https://t.co/DuunFOEZR8 https://t.co/FtCben99QU

2022-06-06 19:57:28 So proud of my dear, amazing friend @ClaireWyart whose creativity knows no end! https://t.co/rIylIPc3vm

2022-05-28 02:55:19 RT @Plinz: Most underappreciated avenue to practical life extension?

2022-05-27 19:32:23 @tyrell_turing who will take signed reviews personally, who are inept at generating social media interest, etc.  While the current system has many faults, I like that the reviewing system insulates us as much as possible from social contact.  I for one would like to keep it that way.

2022-05-27 19:32:04 @tyrell_turing I agree. Also, I feel like a lot of these discussions about changing publishing (ppr, open review) are driven by folks with zero social fear. The proposed changes may not work well for those of us on the other end of the spectrum,

2022-05-27 12:25:25 For young scientists who dream of doing the best science in the world while also changing how science is done for the better: https://t.co/CUBNxW7EHA

2022-05-26 21:44:22 @neurograce @constan_lab @nyuniversity @NYUPsych @NYUDataScience Congratulations, Grace and NYC neuro! So happy to hear this!

2022-05-20 08:11:00 CAFIAC FIX

2022-10-27 15:53:48 RT @Nancy_Kanwisher: Marge's work is hugely important scientifically and clinically, answering fundamental questions about brain developmen…

2022-10-27 15:50:39 I have seen how deeply and personally Marge cares for the animals in her lab, she is truly the last person who should be targeted. https://t.co/LgivwWDFTE

2022-10-26 19:02:31 On cloud 9 after reading this beautiful, brilliant perspective by Qasim Zaidi on 3d vision. An homage to Poincare’s insight that the origin of 3d vision is the isomorphism between the groupoid of transformations between perspectives &

2022-10-25 19:14:21 @tyrell_turing @NSERC_CRSNG Congrats Blake!!!

2022-10-21 23:17:53 https://t.co/DbWFqFnPR4 https://t.co/5597bH5ojO

2022-10-29 14:57:06 @gordonneuro @ndosenbach @DrDamienFair @GrattonCaterina @DeannaJGreene @smarek0502 @RoselyneChauvin @BarchDbarch @ChadMSylvester @DrAshRajesh @dillannewbold Very interesting, congratulations! Underscores the dire need for full human and macaque projectomes--will unleash so many discoveries.

2022-10-27 15:53:48 RT @Nancy_Kanwisher: Marge's work is hugely important scientifically and clinically, answering fundamental questions about brain developmen…

2022-10-27 15:50:39 I have seen how deeply and personally Marge cares for the animals in her lab, she is truly the last person who should be targeted. https://t.co/LgivwWDFTE

2022-10-26 19:02:31 On cloud 9 after reading this beautiful, brilliant perspective by Qasim Zaidi on 3d vision. An homage to Poincare’s insight that the origin of 3d vision is the isomorphism between the groupoid of transformations between perspectives &

2022-10-25 19:14:21 @tyrell_turing @NSERC_CRSNG Congrats Blake!!!

2022-10-21 23:17:53 https://t.co/DbWFqFnPR4 https://t.co/5597bH5ojO

2022-11-17 21:00:21 Interested in helping run a lab to understand visual perception in primates? Tsao lab @UCBerkeley is looking for a new lab manager. Perfect for highly motivated candidates who want to gain deep research experience before grad school. Experience with NHPs is a plus. Please RT

2022-11-16 15:13:54 @ChrizKlink @paolo_papale @KohitijKar Thank you so much, Chris, this means a lot coming from you!

2022-11-16 15:13:07 @SaberaTalukder @SfNtweets Thank you so much, Sabera!

2022-11-17 21:00:21 Interested in helping run a lab to understand visual perception in primates? Tsao lab @UCBerkeley is looking for a new lab manager. Perfect for highly motivated candidates who want to gain deep research experience before grad school. Experience with NHPs is a plus. Please RT

2022-11-16 15:13:54 @ChrizKlink @paolo_papale @KohitijKar Thank you so much, Chris, this means a lot coming from you!

2022-11-16 15:13:07 @SaberaTalukder @SfNtweets Thank you so much, Sabera!

2022-11-18 04:17:13 @UrnovFyodor Fyodor, I am so touched!

2022-11-17 21:00:21 Interested in helping run a lab to understand visual perception in primates? Tsao lab @UCBerkeley is looking for a new lab manager. Perfect for highly motivated candidates who want to gain deep research experience before grad school. Experience with NHPs is a plus. Please RT

2022-11-16 15:13:54 @ChrizKlink @paolo_papale @KohitijKar Thank you so much, Chris, this means a lot coming from you!

2022-11-16 15:13:07 @SaberaTalukder @SfNtweets Thank you so much, Sabera!

2022-11-18 04:17:13 @UrnovFyodor Fyodor, I am so touched!

2022-11-17 21:00:21 Interested in helping run a lab to understand visual perception in primates? Tsao lab @UCBerkeley is looking for a new lab manager. Perfect for highly motivated candidates who want to gain deep research experience before grad school. Experience with NHPs is a plus. Please RT

2022-11-16 15:13:54 @ChrizKlink @paolo_papale @KohitijKar Thank you so much, Chris, this means a lot coming from you!

2022-11-16 15:13:07 @SaberaTalukder @SfNtweets Thank you so much, Sabera!

2022-11-18 04:17:13 @UrnovFyodor Fyodor, I am so touched!

2022-11-17 21:00:21 Interested in helping run a lab to understand visual perception in primates? Tsao lab @UCBerkeley is looking for a new lab manager. Perfect for highly motivated candidates who want to gain deep research experience before grad school. Experience with NHPs is a plus. Please RT

2022-11-16 15:13:54 @ChrizKlink @paolo_papale @KohitijKar Thank you so much, Chris, this means a lot coming from you!

2022-11-16 15:13:07 @SaberaTalukder @SfNtweets Thank you so much, Sabera!

2022-11-18 04:17:13 @UrnovFyodor Fyodor, I am so touched!

2022-11-17 21:00:21 Interested in helping run a lab to understand visual perception in primates? Tsao lab @UCBerkeley is looking for a new lab manager. Perfect for highly motivated candidates who want to gain deep research experience before grad school. Experience with NHPs is a plus. Please RT

2022-11-16 15:13:54 @ChrizKlink @paolo_papale @KohitijKar Thank you so much, Chris, this means a lot coming from you!

2022-11-16 15:13:07 @SaberaTalukder @SfNtweets Thank you so much, Sabera!

2022-11-18 04:17:13 @UrnovFyodor Fyodor, I am so touched!

2022-11-17 21:00:21 Interested in helping run a lab to understand visual perception in primates? Tsao lab @UCBerkeley is looking for a new lab manager. Perfect for highly motivated candidates who want to gain deep research experience before grad school. Experience with NHPs is a plus. Please RT

2022-11-16 15:13:54 @ChrizKlink @paolo_papale @KohitijKar Thank you so much, Chris, this means a lot coming from you!

2022-11-16 15:13:07 @SaberaTalukder @SfNtweets Thank you so much, Sabera!

2022-11-18 04:17:13 @UrnovFyodor Fyodor, I am so touched!

2022-11-17 21:00:21 Interested in helping run a lab to understand visual perception in primates? Tsao lab @UCBerkeley is looking for a new lab manager. Perfect for highly motivated candidates who want to gain deep research experience before grad school. Experience with NHPs is a plus. Please RT

2022-11-16 15:13:54 @ChrizKlink @paolo_papale @KohitijKar Thank you so much, Chris, this means a lot coming from you!

2022-11-16 15:13:07 @SaberaTalukder @SfNtweets Thank you so much, Sabera!

2022-11-18 04:17:13 @UrnovFyodor Fyodor, I am so touched!

2022-11-17 21:00:21 Interested in helping run a lab to understand visual perception in primates? Tsao lab @UCBerkeley is looking for a new lab manager. Perfect for highly motivated candidates who want to gain deep research experience before grad school. Experience with NHPs is a plus. Please RT

2022-11-16 15:13:54 @ChrizKlink @paolo_papale @KohitijKar Thank you so much, Chris, this means a lot coming from you!

2022-11-16 15:13:07 @SaberaTalukder @SfNtweets Thank you so much, Sabera!

2022-12-08 14:50:16 @pfau Which is the more "powerful" part of a computer, the power supply or the processor?

2022-12-08 14:50:16 @pfau Which is the more "powerful" part of a computer, the power supply or the processor?

2022-12-08 14:50:16 @pfau Which is the more "powerful" part of a computer, the power supply or the processor?

2022-03-12 08:11:00 CAFIAC FIX 2022-01-17 08:11:00 CAFIAC FIX 2022-01-11 08:11:00 CAFIAC FIX 2021-12-27 08:20:00 CAFIAC FIX 2021-12-17 07:21:08 @hardmaru I would tell myself that the struggle of discovery is all that matters and the rest is just ribbons and noise. It will all average out and be fair in the end. 2021-12-17 06:56:57 @pfau @michael_nielsen Imagining someone writing this on an NIH specific aims page... 2021-12-17 06:48:29 @svoboda314 Wow, congratulations @ruimcosta , and welcome to the west coast! Many blessings on your new adventure! 2021-12-17 06:44:09 @ShamKakade6 @ChanZuckerberg @Harvard @blsabatini Congratulations Sham and @blsabatini! This is so cool! (Sham, don't know if you remember me from college...amazing to see you in this new role ) 2021-11-14 03:17:28 RT @DeepMind: Can the visual brain learn to represent faces by disentangling? Today in @NatureComms, our team finds that novel faces can… 2021-11-06 23:20:00 CAFIAC FIX 2021-11-01 19:20:00 CAFIAC FIX 2021-08-19 17:52:40 RT @nanoassembly: Woohoo! We got our first ever grant! Thank you @doescience! We’re looking for an ambitious postdoc interested in assembli… 2021-08-13 19:36:39 @KohitijKar @jeff_schall @J_D_Crawford @LaurieMWilcox1 @CentreforVisio1 @vistayorku Congratulations, Kohitij!!! 2021-08-04 20:42:08 @Labrigger Thanks for the link! We need someone with engineering talents who can come to our lab and help brainstorm with us... 2021-08-04 19:51:09 Can anyone recommend a talented freelance machinist in the LA area? (Caltech had a dozen machine shops when I arrived a decade ago, and now has one extremely overbooked one) 2021-07-30 03:27:24 RT @UCI_cycling: "Nobody expected me to come away with a top result here." Listen to @AnnaKiesenhofer describe what it means to be th… 2021-07-30 00:51:33 A truly special phd thesis. https://t.co/65XwxjtTfW 2021-07-23 02:55:12 @meganakpeters @BrianOdegaard2 @UCBerkeley @neuromatch Thanks so much for the tip, Megan! 2021-07-23 00:28:41 Interested in helping run a lab to understand visual perception in primates? The Tsao lab @UCBerkeley is looking for a new lab manager. Vet tech and/or previous NHP experience a plus, but also open to other highly motivated candidates. We do ephys, fMRI, and much more. Please RT 2021-07-21 22:37:09 @BlavatnikAwards @kaymtye @salkinstitute Congratulations, Kay!!! 2021-07-08 04:54:03 @Neurocircuits LOL! I feel more like this: https://t.co/1XaiZcov16 2021-07-07 19:33:18 @ishan276 @attractor_Lu @HesseJanis @joe_wekselblatt @FrankLanfranchi We wondered about this too, and thus tested if mice could detect a square moving at exactly the same speed, but with background stationary (Fig. S1a). Since mice could do this without problem, they were definitely not blind to the two motions present in the stimulus per se. 2021-04-24 21:07:55 Incredible discovery from Lea Goentoro. Also, best first sentence ever: "In contrast to humans’ poor ability to regenerate, the animal world is filled with seemingly Homeric tales: a creature that regrows when halved or a whole animal growing from a small body piece." https://t.co/nKPKtlyvEE 2021-04-24 17:39:14 RT @QuantaMagazine: Scientists have built deep neural networks that can map between infinite dimensional spaces. https://t.co/NvKvolfLcR ht… 2021-04-23 11:49:04 @hakwanlau Congratulations to you and to Riken! Hope we can celebrate in person before you cross the ocean. 2021-04-13 18:46:05 @TimKietzmann @DondersInst @AI_Radboud Congratulations Tim! 2021-04-07 02:09:36 @nicolehcoates @mitbrainandcog Wow, congratulations Nicole! I am so happy for you!!! 2021-04-05 19:53:28 RT @USBrainAlliance: ICYMI, @ScienceMagazine story about functional ultrasound neuroimaging (funded in part by BRAIN Initiative grant to @m… 2021-04-01 17:06:45 @neurograce In https://t.co/EHIwVko1sx we show one reason why: vgg-face, a network trained on face identification, is much more illumination invariant than neurons in face patch AM. The goal of IT does not seem to be simply invariant face identification. 2021-03-31 01:55:48 @strangetruther https://t.co/IBujvDU4DD Try googling van den broeke autoencoder thesis pdf? 2021-03-31 01:17:45 This Master's Thesis by Gerben van den Broeke on generative models and how the brain discovers structure in the world is amazing. It's the most lucid, inspiring, joy sparking piece of scientific writing I've read in a long while. https://t.co/IBujvDU4DD 2021-03-29 22:14:11 @madsarv @EliseSavier @RafaelGrytz @AErisir Huge thank you for organizing this great meeting! I still remember the sfn symposium org. by @cris_niell and @hubermanlab that launched the field of mouse vision. I suspect we'll look back at today's meeting similarly a decade hence, as a milestone day for tree shrew neuroscience 2021-03-23 13:32:28 @JSDaniels97 @PsychPrinceton Congratulations, Jonathan! So happy to hear this wonderful news! (though sad I won't see you at Berkeley) 2021-03-22 01:06:09 @Neurocircuits Thanks so much for reading and for sharing your memory, Marc! 2021-03-17 03:33:06 When will the darkness lift? Heartbroken over the senseless death of my sisters https://t.co/f7kGpm5Rfj 2021-03-14 15:24:29 @AthenaAkrami @Yuelin_Shi @NeuroTheoryUCSD @StefanoFusi2 Thank you so much, Athena! That means a lot coming from you 2021-03-13 03:33:18 (16/16) Reaching this understanding is a testament to the incredible ingenuity and persistence of Liang She (who is on the job market) and the deep insight of our collaborators Marcus Benna @NeuroTheoryUCSD and @StefanoFusi2. 2021-03-13 03:32:53 (15/16) Distilling all of these findings, here is our current geometric picture of how face memory works: https://t.co/x79pl3jJtC 2021-03-13 03:32:31 (14/16) Besides our central finding of familiarity-gated axis change, we were surprised by how prominent axis tuning for unfamiliar faces was in perirhinal cells. This suggests that the perirhinal object code is not yet explicitly semantic. https://t.co/OajGz8MYyH 2021-03-13 03:32:05 (13/16) Silencing PR did not alter the long-latency axis change in AM for familiar stimuli. So face memories are either intrinsic to IT or arise from interactions with areas outside the medial temporal lobe. https://t.co/txQ7QC0tQe 2021-03-13 03:30:53 (12/16) Instead, we could decode familiarity from a short-latency shift between the familiar and unfamiliar neural coding spaces that occurred before any difference in mean response to familiar vs. unfamiliar faces. 2021-03-13 03:30:34 (11/16) Indeed, we found that repetition suppression depends strongly on temporal context, making it a non-robust signal for familiarity. https://t.co/vtmGlCMe2f 2021-03-13 03:30:00 (10/16) Our finding that familiarity leads to long-latency axis change departs from the widely-accepted notions of repetition suppression and monotonic sparsening. https://t.co/vztg2XLJrM 2021-03-13 03:29:39 (9/16) This explains a striking fact about human face perception: we can determine the identity of familiar faces much better than unfamiliar ones. Here is my favorite demo of this: https://t.co/53ph4vALHw 2021-03-13 03:29:11 (8/16) Since the feature distributions were matched, you shouldn’t. But AM and PR neurons differentiate the two sets, and distinguish the familiar faces much better than unfamiliar! https://t.co/pJHmstoIev 2021-03-13 03:28:37 (7/16) At the same time, the neural distance between familiar faces was higher than that between unfamiliar faces. Below are two sets of faces, one familiar and one unfamiliar to the monkey. Do you notice any big difference between the two? https://t.co/eTA5IJPWIs 2021-03-13 03:28:13 (6/16) So if you use a decoder trained on unfamiliar faces to decode familiar faces, it works well at short latency. But it completely fails at long latency: https://t.co/kvXhezYCzQ 2021-03-13 03:27:30 (5/16) TLDR: We found that the code is indeed changed by familiarity. Our central finding is that in face patches AM and PR, familiar faces are encoded by a *different* axis compared to unfamiliar faces at long latency. https://t.co/BHhqMiMdvp 2021-03-13 03:26:54 (4/16) This made us wonder, does this code change in any way for familiar faces? If faces are sensed and remembered by the same cortical cells, then it seems this face code should be altered by familiarity. 2021-03-13 03:26:29 (3/16) Three years ago, we reported the shape-appearance axis code for faces in IT face patches (Chang & 2021-03-13 03:26:06 (2/16) A central assumption of neuroscience is that long-term memories are stored in cortex. But we know surprisingly little about the nature of these memories. https://t.co/a20gTgbAUJ 2021-02-10 18:32:15 This would be a great stimulus to localize the humor area(s) of the brain. Condition A: soundtrack alone Condition B: video alone Condition C: both (Uncle Howard: don't worry, I'm not planning to do this with your money :)) https://t.co/rEzfS2JUSH 2021-02-10 15:03:39 RT @HiroTaiyoHamada: CC Tokyo 2/25th 17-19 (PST) Happy to announce the next CC talk by Dr. Janis Hesse @HesseJanis at Caltech @doristsao… 2021-01-26 12:25:19 @kyo_iigaya @ColumbiaPsych @Columbia Congratulations, Kyo! 2021-01-21 19:31:46 Tirin Moore @StanfordBrain @HHMINEWS will receive the 2021 NAS Pradel Research Award recognizing his groundbreaking contributions to understanding visual attention! Congrats Tirin!!! https://t.co/1m8NknCokw 2021-01-21 19:21:00 Huge congratulations to the one and only Nicole Rust @VisualMemoryLab on the 2021 NAS Troland Research Award! Nicole brings such elegance, computational rigor, and creativity to understanding the wilderness of visual memory https://t.co/X0otjBxXCd https://t.co/8ENvnSav1F 2021-01-15 21:21:51 @francesarnold @eric_lander @maria_zuber What wonderful news! Congratulations and thank you, Frances! 2021-01-15 18:09:12 @CountyofLA @lapublichealth I don't understand the logic. Less than 33% of available doses have been given in LA County = much too slow. Eligibility was lowered to 65+ to increase the speed (just like turning up water pressure makes water come out of hose faster). Does LAPH not follow? 2020-12-24 00:19:43 @rkarmani @bayesianboy Yes, the idea of predictive coding as biologically plausible model of network learning is described beautifully by Lillicrap et al in https://t.co/ACv858iXN9. Amazingly, the idea of predictive coding was already present in McCulloch and Pitts 1943: https://t.co/uO2KAD1orH 2020-12-22 17:03:33 RT @GaryMarcus: Topics for Wedesday's all-star #AIDebate2! featuring @yudapearl @drfeifei @mmitchell_ai @YejinChoinka @rcalo @kenneth0st… 2020-12-18 04:50:01 @c_constan Congratulations Christos and Vanderbilt! 2020-12-17 22:52:53 @BenPurkert See Jewel's elegy for Tony Hsieh, where she describes his definition of success: being willing to risk everything (fail) because you believe in an idea so much. Last line is perfect, letting son to know he was meant to fly. 2020-12-16 16:48:39 https://t.co/tjOjdubRNe 2020-12-15 17:05:44 They inspire us, and we feel bereft when they’re gone. Leslie was such a person. 2020-12-15 17:05:18 There are certain rare individuals in science who care vitally not just for their own next discovery, but for the progress of the field as a whole, the struggles and successes of all those who love the same problem they love. All of us, near or far, can sense their presence (1/2) https://t.co/u0U1bHMACe 2020-12-15 15:22:33 @MatthiasMichel_ ...and it's true for all illusions: they're not illusions in the sense that they result from a concrete neural process that is trying to model a larger reality. As Freud generously said, 'behind every delusion is a kernel of truth.' 2020-12-15 15:06:59 @MatthiasMichel_ Very interesting, look forward to your piece! I like the analogy to amodal completion. I think this illusion of detail is, like amodal completion, not an illusion. I suspect it's based on an invariant tracking mechanism. Once you've seen sth sharply, it stays in your world model. 2020-12-12 01:22:14 @casa_tuthill @eabuffalo @UW What a wise and brilliant choice! Congrats Beth and @UW Physiology! 2020-12-12 01:20:46 @MillerLabMIT This is so sad.The concept of dorsal vs ventral stream is truly profound, one we are just beginning to grapple with. She was very kind to me. 2020-12-11 19:27:24 @weijima01 @Slate So cool to see your byline in Slate! I had to do a double take :) ...thank you for teaching me the beautiful phrase 'radically egalitarian' (through your specific use of the term and through the spirit of your whole piece) 2020-12-11 14:02:57 A super cool and unexpected finding about retinotopy in a species that a lot more vision scientists should be studying https://t.co/WvKeTYtPqj 2020-12-09 16:12:56 Looking forward to this (and I'd look forward even to being a fly on the wall with these human intelligences) https://t.co/z1RRBEBvJ3 2020-12-08 19:37:49 RT @YujiaHuangC: I will be presenting our work: Neural networks with recurrent generative feedback tomorrow (Wed) @ 09:00 – 11:00 PST. We s… 2020-12-08 14:17:01 @shadlen I'm so sorry to hear this. Not knowing anything except that she was your mom, I know she must have been an extraordinary human being. Sending you love through the darkness. If I can help in any way, I'm here for you. 2020-12-05 21:05:33 @thcindywu Looks amazing, I'll be there soon! 2020-12-04 03:01:13 RT @BenPurkert: this poem contains the whole world inside it Garous Abdolmalekian trans. by Ahmad Nadalizadeh and Idra Novey https://t.co… 2020-12-04 00:26:13 @schneiderneuro this is a great example of the part whole effect! (the newly revealed bottom half of the face is interfering with your perception of the top half of her face) https://t.co/FrXb8YQR6d 2020-12-04 00:21:49 @schneiderneuro Interesting! It sounds like the presence/absence of mask on your kids has become a lot less salient. I suspect this has to do with IT -&gt 2020-12-03 01:36:42 A wonderful offer of assistance to young women from @pollyp1! https://t.co/8Nh0sKLlOn 2020-11-27 01:14:28 Staring at this picture in awe after reading this wonderful piece by @EricaKlarreich https://t.co/rWnBKTTP5H https://t.co/cqA7ZqnNqw 2020-11-21 19:20:05 @pollyp1 @UCBerkeley @opheliavosshall Thank you so much, Leslie! Wow, I look forward to meeting Ophelia! 2020-11-21 19:12:02 @pollyp1 Sometimes I think it would be best if we were all prosopagnosic... https://t.co/puPKQCH8j8 2020-11-19 14:32:40 @IshmailSaboor @ZuckermanBrain @Columbia_Bio Congratulations to you and to Columbia!!! 2020-11-18 16:35:36 RT @MariaSpiropulu: Madam Wu and her weak interaction parity violation experiment was Nobel worthy and crucial for the development of th… 2020-11-18 14:30:14 RT @Antihebbiann: The Kennedy lab is recruiting a computational technician to help develop our lab's behavior analysis software! Apply at h… 2020-11-17 15:25:47 @behrenstimb @KordingLab I think it's fair to say that in general, when anyone sedentary scoffs at anyone struggling, it doesn't turn out well for the former 2020-11-17 01:26:06 @joeyginorio @KordingLab If there is a system comprised of discrete steps a psychologist might want to know what these steps are? (certainly very hard to figure out using just behavior, but one could try masking at different delays etc) 2020-11-16 23:32:10 @KordingLab Location tells a lot! Fact that face patches were located in similar places across hemis and animals was 1st clue they were not just random columns of face cells, but a system. Fact they were so spread out suggested neighboring cortex doing similar computations for other objects. 2020-11-16 14:44:58 Matthias Michel has written a very thoughtful essay on our recent paper on binocular rivalry. His consciousness blog has wonderfully lucid discussions about many exciting and deep topics (including two of my favorite, object files and postdiction). https://t.co/whtEhp0n4X 2020-11-15 17:12:26 @neuroecology So experience seems to be necessary to tune up neurons within a region that is already pre-disposed to become face selective extremely early on (either due to just 1 month of experience w/ face &amp 2020-11-15 17:06:10 @neuroecology My understanding of Marge's results is more the opposite: face-deprived monkeys *don't* preferentially look at faces (compared, e.g., to hands). Also, the location of areas that become face patches are determined from 1 month (as early as she could scan) (Fig 3, 4, NC 2017). 2020-11-15 02:04:34 @hollyshortall People who use the words semantic, bootstrap, normative, or basal ganglia with ease 2020-11-13 13:30:39 @MintakaGlow My senior year of HS, I was so nervous about how hard classes @Caltech would be, I started reading the Feynman Lectures. That summer of self study turned out to be more meaningful than any class I took afterwards. 2020-11-11 23:29:47 Whatever the answer, the truth is always the most interesting. 2020-11-11 23:29:29 What does all this mean? We think our results are compatible w/ 2 schemes. Cells could show true mixed selectivity for the conscious and suppressed stimulus (left) or it could turn out that some spikes represent the conscious stimulus, others the physical stimulus (right—my bet). https://t.co/50rXb770gi 2020-11-11 23:28:10 This is one of those cases where storytelling got ahead of actual data. The original paper by Sheinberg and Logothetis (1997) clearly showed that modulation in IT by the conscious percept was weaker than by the physical stimulus. https://t.co/3o3nIYbytI 2020-11-11 23:27:34 Thus the classic cartoon picture of binocular rivalry in IT cortex is wrong. IT cells multiplex both the conscious percept and suppressed stimulus. What’s more, the degree of modulation by the two is correlated in single cells. https://t.co/CLE5U191yY 2020-11-11 23:17:07 When he changed the identity of the suppressed stimulus during binocular rivalry, and compared (A, B) versus (A, C), he could decode the suppressed stimulus B versus C with 74% accuracy. https://t.co/B58rSz2JwE 2020-11-11 23:16:42 In the just published paper in @eLife, Janis presents a critical new piece of evidence that IT cells don’t just represent the conscious percept. 2020-11-11 23:15:51 Below are histograms of single trial responses from face patch ML to physical and perceptual alternations, labeled by the percept. Notice how different the distributions are: https://t.co/xkS12A2pjO 2020-11-11 23:15:09 Our biorxiv preprint already presented an important piece of evidence casting doubt on this notion. https://t.co/FNpZGYA6oW 2020-11-11 23:13:44 An oft-told story in the field of consciousness studies is that activity in V1 reflects the physical stimulus while activity in IT cortex reflects only what is consciously perceived. It’s an appealing story, but @HesseJanis shows it’s not true. https://t.co/gyOjPsSc9G 2020-11-11 18:18:12 In this crazy year, this documentary is one of the most beautiful things I experienced. https://t.co/dRSOcOXSMl 2020-11-11 13:41:29 RT @QuantaMagazine: Quanta’s spring and summer interns work alongside staff editors and reporters to report and write about fundamental dis… 2020-11-10 01:42:16 RT @guptaviolin: The look in her eyes. https://t.co/IOBxFGXupP 2020-11-08 20:39:18 RT @nanoassembly: The fact that the election is this close tells us that we need to work harder and help more Americans embrace logic, scie… 2020-11-06 14:54:10 RT @CaltechN: Learn more about the Chen Institute BrainWAVE fellowship program, for summer research in neuroscience, during the information… 2020-11-04 21:56:51 RT @nanoassembly: I really hope I can continue interviewing non-US candidates! They are the landslide majority of applicants to my lab. The… 2020-11-04 17:00:37 @BateupLab @berkeleyMCB @UCBerkeleyNeuro Congratulations, Helen!!! 2020-11-03 18:29:14 I wrote a review with @HesseJanis about the macaque face patch system. Apologies to those who find the title gruesome, it's meant to convey how the face patches make seemingly inaccessible questions about the brain tractable https://t.co/tIjhqBmUoH 2020-10-30 03:41:33 Love https://t.co/yEEx8CVEl9 2020-10-29 15:47:09 @dabacon @IonQ_Inc @Kellogg_Girl Congrats Dave!!! (don't know if you remember me from undergrad days...) 2020-10-28 16:35:21 @SfNtweets @VivianGradinaru @Caltech @TheHongLab @UCLA @Sunovion Wow, what a great choice! Congratulations @VivianGradinaru and @TheHongLab! So happy for you! 2020-10-27 20:30:44 RT @SumnerLN: (1/N) Have you ever wondered how your #brain represents the meaning of language? Despite the centrality of language to the hu… 2020-10-27 20:10:37 Thank you for reading my essay, @JoyceCarolOates. This means more to me than you can imagine. https://t.co/yaAFhNilSO 2020-10-27 15:06:50 It's part of a special issue dedicated to Charlie organized by Michael Graziano and @MillerLabMIT, with many wonderful pieces, including this essay by @JoyceCarolOates https://t.co/Rl4cO8YN13 2020-10-27 15:06:49 I wrote an essay in honor of Charlie Gross. It's a celebration of the lone explorer, a story about the ups and downs of scientific fortune, and a reply to the perennial argument that recording from single neurons doesn't tell us much about the brain. https://t.co/ZeywN34Td7 2020-10-27 06:12:39 @brainpicker At your sixtieth birthday, in the cake’s glow, Ariel sits on your knuckle. You feed it grapes, a black one, then a green one, From between your lips pursed like a kiss. ... What about Ariel? Ariel is happy to be here. Only you and I do not smile. TH 2020-10-26 02:38:28 I know this is gonna be so special, go if you're within driving distance of Monterey and you love classical music! https://t.co/c5QzPaE0vG 2020-10-26 02:32:41 @shadlen Elaborate please? 2020-10-25 12:18:31 @StanDehaene I'm the guilty party. I apologize for the implication that you're only responsible for this one arrow :) 2020-10-24 18:44:20 @neuro_data @behrenstimb The book 'Shadows of the Mind' 2020-10-24 13:48:04 @tiagogmarques Shadows of the Mind 2020-10-24 13:35:38 I would argue we don't even need psychology to understand a large part of the structure of mental representations. We all have an intimate window into it as conscious introspective human beings. 2020-10-24 13:35:37 The figure below by Roger Penrose is perhaps my favorite figure of all time. It so powerfully and inspiringly rejects the notion of 'primacy' and embraces the notion of a single mystery to be solved. https://t.co/8Lw2G5YWPJ 2020-10-24 13:28:35 @yvetteefisher @UCBerkeley @UCBerkeleyNeuro @berkeleyMCB Same here, Yvette! 2020-10-24 03:04:12 @UeliRutishauser @UCBerkeley @UCBerkeleyNeuro @berkeleyMCB @Caltech That would be amazing! :) 2020-10-23 21:40:04 RT @christineliuart: So....when I was applying to grad school I never thought I'd end up at a place like @UCBerkeleyNeuro. Shoot your shot… 2020-10-23 16:52:57 Attest to this 100%, Joe is brilliant and I would love to have him as a faculty colleague. https://t.co/Pv4RMiRTH5 2020-10-22 22:26:39 @schoppik @KernelCo My understanding from Q&amp 2020-10-22 22:05:35 Loved the futurist vision shared by @KernelCo. Their new fNIRS system maps the brain at 1 cm &amp 2020-10-22 02:48:36 Looking forward! https://t.co/1vb9yyTHaS 2020-10-20 20:05:54 Wow, congratulations @MichelMMaharbiz and @blancinegre1972! https://t.co/vmwdXNv2KT 2020-10-20 13:21:20 @patchurchland @NautilusMag @wtgowers makes the same point about the importance of fluency in understanding math in the wonderful little book 'Mathematics: A Very Short Introduction,' and suggests learning by rote is actually a more abstract categorical approach to understanding math https://t.co/izYafCCefr 2020-10-19 16:16:03 @russpoldrack Yes, we'll be neighbors again! 2020-10-19 16:08:21 I am excited to announce that my lab will be moving to @UCBerkeley in fall 2021. Grateful to join the incredible neuroscience community @UCBerkeleyNeuro @berkeleyMCB and become part of an institution that is truly disruptive, scientifically and culturally. https://t.co/VDnGSeUOUn 2020-10-14 17:10:27 RT @mikhailshapiro: "What can ultrasound do for me?" - our neuroscience colleagues (you know who you are). Thanks @NeuroCellPress for giv… 2020-10-13 03:22:17 @artistzurbaran @raj_raizada Raj! This is the inspiration for Morten Lauridsen's O Magnum Mysterium. Maybe that's how you found it too? Come to Pasadena and we can go see it together at the Norton Simon. 2020-10-09 23:42:52 RT @CaltechN: Applications open for BrainWAVE fellowships which offer opportunities for undergraduates interested in carrying out a summer… 2020-10-08 05:24:27 RT @VamsiMootha: CRISPR is just different — it towers so large, transcends science, so important for all of society and our future. Rarely… 2020-10-08 02:30:38 Yes! Here's a glimpse into what it would take: https://t.co/xe4A6Y2QBQ https://t.co/uutjTOzgaR 2020-10-06 19:14:40 Congratulations to my amazing colleague Magda Zernicka-Goetz, a true pioneer! https://t.co/YsQy1csjDq 2020-10-06 18:29:33 RT @nanoassembly: By analyzing the deviations from their trajectories we can calculate the position of the black hole https://t.co/agmc0B1h… 2020-10-06 16:49:17 @preskill August 14, 2018 in Cahill during his visit to Caltech 2020-10-06 15:55:23 So happy to hear that the hero of my youth, Roger Penrose, has been awarded the Nobel Prize. His honesty in stating that existing physical laws can't explain consciousness influenced me hugely. I used to daydream about being his grad student. https://t.co/e3XI9Ya4Qg 2020-10-03 13:07:17 @Caltech As someone who considers Millikan to be a home of sorts, I hope we can rename it already. The truth shall set you free. 2020-10-03 13:07:06 @Caltech In her extraordinary letter, Sarah Sam shared with all of us the suffering caused by the Task Force's deliberations. Her integrity &amp 2020-09-30 22:01:48 @SohailTavazoie @BernardoTavora Wow. So far outside of my field, but this sounds absolutely incredible, congrats Sohail! 2020-09-26 02:54:08 @casa_tuthill @KakaniKatija @USFigureSkating @UWaeroastro @UW LOL. Did I really claim this? There is a thing called adult figure skating. It's to figure skating what watching star trek is to being an astronaut. I'm an adult figure skater. 2020-09-26 02:41:22 @DeanMobbs Congrats Dean! So happy to hear this, we need to celebrate! 2020-09-25 00:00:58 RT @apvadaparty: I’m a recent graduate of Johns Hopkins (B.S. Neuroscience, B.S. Computer Science) with a year of industry software enginee… 2020-09-13 13:22:34 Planet earth is red, I'm so glad I left. -Major Tom, 2020 https://t.co/sNHjYHgJ5T 2020-09-13 12:43:49 @cobarnes27 @Stanford @Stanford_ChEMH Congratulations Christopher!!! 2020-09-11 20:12:55 RT @nanoassembly: Happy to see some relief for underserved students: @UCBerkeley @Berkeley_EECS don’t require GREs this year! I'll be happy… 2020-09-11 16:56:27 Huge kudos to @ChrizKlink, Igor Kagan, and @cipetkov for leading a team of over two dozen NHP experts to distill their collective insights into methods for combining perturbation and imaging in NHPs. A highly valuable resource for the NHP community https://t.co/7hJKDTKLLn 2020-09-11 15:03:27 RT @dileeplearning: Interested in the functional logic of cortical and thalamic microcircuits? Check out our new preprint from @vicariousai… 2020-09-11 13:55:06 This beautiful &amp 2020-09-10 16:44:07 What wonderful news, congratulations Catherine @DulacLab for your amazing work and selfless leadership! https://t.co/9zsrDKL1d1 2020-09-07 17:13:32 RT @AdamWierman: Even though the ad has been live for a little bit, I'm new to twitter so I'll post it now: We'll be searching for faculty… 2020-09-07 14:58:04 RT @nanoassembly: Shroedinger’s Grief: When you’re not sure whether your cat who hasn’t come home for three days has been eaten by coyotes,… 2020-09-06 01:53:09 @AnimaAnandkumar @Caltech Thank you so much, Anima! Fun discussing target propagation with you last night :) 2020-08-29 16:32:57 @KordingLab @dileeplearning @behrenstimb @danilobzdok @LucinaUddin @tyrell_turing @Botvinick I spent a large chunk of my PhD recording from disparity-tuned simple and complex cells. The main conclusion was that H&amp 2020-08-29 16:15:30 @KordingLab @dileeplearning @behrenstimb @danilobzdok @LucinaUddin @tyrell_turing @Botvinick It explains the mechanism at an algorithmic level. How the contrast-sensitive, position-sensitive subunits are implemented is a distinct (and IMO, less interesting) question. Realizing complex cells are summing simple-cell like subunits was a huge mechanistic insight. 2020-08-29 15:44:57 @KordingLab @dileeplearning @behrenstimb @danilobzdok @LucinaUddin @tyrell_turing @Botvinick I strongly disagree that the H&amp 2020-08-29 15:16:55 @EricMTrautmann Agree, Eric! I found the demo hugely inspiring--suggests a new era of neuroscience, where you can just take your animal to the surgery robot to get thousands of electrodes implanted wherever you need. Only massive $$$ and focus could bring this level of engineering. 2020-08-27 22:24:36 @EricMTrautmann @BBRFoundation Yay, congrats Eric!!! 2020-08-15 14:45:09 @francesarnold @NASAPersevere and for redefining the concept of grandmother! :) 2020-08-15 13:23:42 @francesarnold @NASAPersevere Congratulations Frances! thank you for sharing this beautiful picture 2020-08-13 16:43:57 @gershbrain Not a single professor wrote back... 2020-08-13 15:25:24 At 20, I sent an email to a dozen professors across Harvard asking for book recommendations. Today, I want to send an email to all my colleagues asking for tips on how to deal with email 2020-08-12 23:30:22 @nanoassembly I think it's the same as picking an ideal collaborator https://t.co/Wz9rEsriWq 2020-08-12 15:41:03 @neurograce @sigmascience Congratulations^2, can't wait to read your book! 2020-08-10 14:14:05 @behrenstimb @TrackingActions @neuroecology @cris_niell Maybe. See this thread for evidence that a famous blindsight patient actual has very poor but conscious vision. https://t.co/StkmCBsf7q 2020-08-07 18:48:39 Congratulations to my brilliant and visionary colleague @VivianGradinaru! https://t.co/qq6DeM1Fx3 2020-08-05 14:06:25 @_fernando_rosas @WiringTheBrain My colleague Matilde Marcolli is grappling with the mystery of qualia by using category theory to consider topological invariants of the set of all possible interpretations (“decodings”) of a neural network. https://t.co/r2YCYRTEB6 2020-08-05 14:05:48 @_fernando_rosas @WiringTheBrain The brain is certainly not just a falling rock, unless falling rocks have qualia. Qualia are the ultimate form of information decoding, completely determined by neural state, requiring no prior assumptions about decoding of an external world. 2020-08-04 15:45:01 Thank you, Kafui, for showing all of us the way. https://t.co/SqyO4FTnDZ 2020-08-02 22:46:23 Did you know that @AstroBehnken's PhD advisor was @Caltech's own Richard Murray, incoming chair of Biology &amp 2020-08-01 02:42:21 @VivianGradinaru @davidvanvalen @Caltech Wow, congratulations David! 2020-08-01 01:56:23 @AnimaAnandkumar @Caltech @anqi_liu33 @kazizzad @YujiaHuangC @SahinLale @JamesGornet @jiawzhao @yyuan_yon @francesarnold @yisongyue @tfmiller3 @lpachter @bethanyehlmann @rmichaelalvarez @davidvanvalen We are all so lucky to have you at Caltech, Anima! Very excited about our collaboration. 2020-07-30 17:42:33 Another wonderful colleague who wishes to remain anonymous has stepped up and offered to provide a 6K Rising Star in Neuroscience Scholarship to a young Black scientist with big dreams. I am so happy to announce the winner, Jonathan Daniels @JSDaniels97 @BlackInNeuro #blackinstem https://t.co/qQFjzPXUrG https://t.co/JMxGDbPPSO 2020-07-29 21:31:40 RT @nanoassembly: Excited to announce that I will start my lab @UCBerkeley @Berkeley_EECS in 2021! My lab will combine molecular self-assem… 2020-07-29 20:06:43 RT @AbrahamYule: Very proud and thrilled to work with Isaias as his McNair faculty mentor. Lots of exciting and promising work to follow! h… 2020-07-29 18:51:18 Breaking news: a wonderful colleague which wishes to remain anonymous has decided to pitch in 6K to fund a second scholarship! I'm delighted to announce the winner, Nicole Coates, who hopes to pursue a PhD in cognitive science @BlackInNeuro https://t.co/EOuTru1xLa https://t.co/LPD3w6bM0l 2020-07-29 16:04:17 I’m delighted to announce the winner of the scholarship, @ghezae_isaias from UCSB. His early experience with injustice has taken him on a thrilling intellectual path. Read his inspiring story &amp 2020-07-29 15:51:14 To all the young Black men and women who sent me essays: I’m profoundly grateful. It’s incredible how you've overcome all barriers to enter Neuroscience, driven by curiosity &amp 2020-07-24 19:27:36 Sad to learn about the passing of Mark Konishi. A humble son of two silk weavers from Kyoto, he came to this country to study how birds learn to sing, and gained some of the first glimpses into the beauty and precision of the neural circuits that give rise to behavior. https://t.co/I1yZ4Ld19m 2020-07-23 17:18:49 RT @DesrochersLab: NHP Research in Danger! Important, this one is flying under the radar (on purpose) - for all those working with animals… 2020-07-17 14:23:00 RT @YujiaHuangC: Presenting our work: Neural networks with recurrent generative feedback at #ICML2020 UDL workshop (7/17, 9-10AM, PDT). Jo… 2020-07-16 18:22:56 @FiebelkornIan @URNeuroscience Congrats Ian!!! So happy for you 2020-07-13 16:16:41 @mjaztwit Congrats Mehrdad!!! 2020-07-13 15:00:56 @neuroecology PS but there's another, more interesting answer to your question, which I hope we can share in a preprint soon 2020-07-13 14:56:49 @neuroecology I think that the number of latents needed for good face representation is on order of 100 (Google FaceNet uses 128 dimensions), and number of neurons in a face patch is on order of 100,000 (1 cubic mm of cortex). So ratio would be 1 latent: 1000 neurons 2020-07-13 14:22:32 @neuroecology Re second question: the paper is not claiming that beta VAE has discovered all the features necessary to represent a face. Indeed another paper analyzing the same data set https://t.co/VkmxGf7qRJ shows that encoding performance increases with additional features &gt 2020-07-10 17:13:16 RT @StanDehaene: I am thinking of starting a new project on data mining of whole-brain fMRI data, searching for basic rules of how mental r… 2020-07-07 21:15:22 @cfcaskeyPhD @VUMCradiology Congratulations Charles!!! 2020-07-07 16:36:29 @neurovium Haha! Yes, this one: https://t.co/j5MnjvEZTy 2020-07-07 16:10:37 @neurovium I will always remember the day he spent with me in Cambridge 2008...sitting in his very beat-up car listening to Mozart...how I wish I could have another conversation with him! This paper touches on almost everything I care about https://t.co/nz9TftnMcB 2020-07-07 13:12:00 RT @neurovium: I learned that Horace (Barlow) passed away. He was a visionary, a neurotheory pioneer. I feel terrible since I had promised… 2020-07-05 00:56:25 RT @meganakpeters: To my friends and colleagues around the world: Many of you know that I have been working with an incredibly passionate a… 2020-07-02 08:26:00 @dileeplearning @behrenstimb Got it, thanks, and looking forward to your new work! 2020-07-02 06:00:54 @dileeplearning @behrenstimb Looks very interesting, thanks!

Découvrez Les IA Experts

Nando de Freitas Researcher at Deepind
Nige Willson Speaker
Ria Pratyusha Kalluri Researcher, MIT
Ifeoma Ozoma Director, Earthseed
Will Knight Journalist, Wired
Plus