retrievealltweets
Il y a 3401 messages d'experts à propos de l'IA
| @attninaction @tyrell_turing @KordingLab @PsychScientists @andpru @neuro_data @jpillowtime @neurograce Totally cool with formalising the word. Just don’t change it into something else (like causality), and don’t make definitions that make our lives harder (because they can’t be tested). | 2021-02-16 18:10:53 |
| RT @SuryaGanguli: 1/ New preprint: | 2021-02-16 18:06:39 |
| RT @Montreal_AI: NeRF−−: Neural Radiance Fields Without Known Camera Parameters Wang et al.: https://t.co/3dqimYovSY #ArtificialIntellige… | 2021-02-16 18:04:57 |
| RT @mer__edith: After firing @timnitGebru + @RealAbril + nurturing a racist and toxic culture, Google has the paternalistic AUDACITY to imp… | 2021-02-16 17:53:33 |
| Nah. We’re just below the bar and need more “training.” https://t.co/vcdgx16mRk | 2021-02-16 17:53:06 |
| RT @itsafronomics: If you're not going to hire us, then why invest in | 2021-02-16 17:52:13 |
| RT @cephaloponderer: Entering week 5 of my manager being locked out of work and week 11 of gaslighting, misinformation spread, and zero acc… | 2021-02-16 17:51:38 |
| RT @emilymbender: Meanwhile, all the way back in early December, @GoogleWalkout laid out the path that @Google should have taken, to make t… | 2021-02-16 17:51:26 |
| RT @emilymbender: And instead of following that clearly articulated path, @Google is promising to | 2021-02-16 17:51:09 |
| RT @garnettachieng: the tech + society resource list i was compiling is here it's not done yet, but so far it has: -open access books -co… | 2021-02-16 17:49:31 |
| RT @KordingLab: The @neuromatch conference paper is finally out: https://t.co/F4LNB1XzPm We learned so much! | 2021-02-16 17:42:57 |
| RT @Montreal_AI: Bayesian Neural Network Priors Revisited Fortuin et al.: https://t.co/T6OO7XETIs #Bayesian #MachineLearning #NeuralNetwo… | 2021-02-16 17:32:45 |
| RT @MelMitchell1: Good article on AI ethics from @NewYorker : https://t.co/S1XW3p3fmv | 2021-02-16 17:32:37 |
| RT @tydsh: Our work https://t.co/NeasPSMKcb analyzes SSL with linear predictor w/o neg pairs. It analytically explains how w decay, predict… | 2021-02-16 17:31:59 |
| RT @madlag: First glimpse of the upcoming Block Sparse Pruning release: 2.44x faster-than-BERT SQuAd v1 network on GPU, with a F1 drop of 0… | 2021-02-16 17:31:54 |
| RT @Montreal_AI: TransGAN: Two Transformers Can Make One Strong GAN Jiang et al.: https://t.co/OmD5k2Vs43 #Transformers #GenerativeAdvers… | 2021-02-16 17:31:49 |
| Very cool idea. I do love the idea that the phylogenetic history of predictive learning in the brain is via energy minimisation cost functions. https://t.co/CzjxbsSV9f | 2021-02-16 17:29:06 |
| RT @TimKietzmann: New preprint alert! We show that predictive coding is an emergent property of input-driven RNNs trained to be energy effi… | 2021-02-16 17:26:24 |
| RT @FAccTConference: ACM FAccT 2021 was slated to take place in Toronto, Canada. However, due to the ongoing COVID pandemic, it will be hel… | 2021-02-16 17:02:40 |
| RT @vincefort: Have you ever wondered whether isotropic Gaussian priors are good enough for your Bayesian neural network weights? They are… | 2021-02-16 17:00:03 |
| RT @FrankNoeBerlin: Online workshop on protein dynamics coming up April 26-28. Awesome speakers including @RommieAmaro @LindorffLarsen @gde… | 2021-02-16 16:58:40 |
| RT @tunguz: I took a A100 for a quick spin. These numbers are insane! I trained a single XGBoost model on the Porto Seguro data. Training o… | 2021-02-16 16:58:25 |
| RT @ziruiwang_ox: Check out NeRF--: NeRF without camera parameters! We show that camera poses, intrinsics and NeRF can be jointly optimised… | 2021-02-16 16:57:42 |
| RT @qlhoest: Introducing Datasets v1.3.0! 600+ datasets 400+ languages load in one line of Python and with no RAM limitations With… | 2021-02-16 16:57:34 |
| RT @FAccTConference: ACM FAccT 2021 was slated to take place in Toronto, Canada. However, due to the ongoing COVID pandemic, it will be hel… | 2021-02-16 16:52:17 |
| @jacobmbuckman @cinjoncin @timnitGebru @mmitchell_ai And | 2021-02-16 16:51:24 |
| @jacobmbuckman @cinjoncin @timnitGebru @mmitchell_ai But there's space between | 2021-02-16 16:50:18 |
| 2021-02-16 16:49:50 |
| RT @JWhitehurst: The #climate crisis is one of the most pressing issues of our time. Today, @IBM committed to net zero greenhouse gas emiss… | 2021-02-16 16:49:15 |
| @jacobmbuckman @cinjoncin @timnitGebru @mmitchell_ai Thanks for the flag. So: I think we might be talking past each other here. There is no such thing as a fully debiased dataset, and I don't think any of the Stochastic Parrots authors would say there is. > | 2021-02-16 16:49:12 |
| Good article on AI ethics from @NewYorker : https://t.co/S1XW3p3fmv | 2021-02-16 16:46:32 |
| Question: How many will @google hire? https://t.co/fQfDjKY2eI | 2021-02-16 16:39:05 |
| RT @vinayprabhu: (2/n) The hagiography of Google's researchers who have peddled peer-reviewed papers based on these opaque datasets reads l… | 2021-02-16 16:18:02 |
| Remember this: — Not every software-intensive system is web centric. — Not every web centric system requires elastic scaling | 2021-02-16 16:11:54 |
| RT @Tesla: Winter at Giga Texas https://t.co/eqL8lPICKe | 2021-02-16 16:00:45 |
| RT @Babak_Taati: | 2021-02-16 15:59:22 |
| @cian_neuro I'm okay with it... It is indeed a reflection of the reality of biological measurements, IMO. | 2021-02-16 15:58:13 |
| RT @UofMAfrica: | 2021-02-16 15:48:48 |
| @dkaushik96 @jeffbigham I’m going to patent a system that uses a drone to bring cars to delivery packages | 2021-02-16 15:21:01 |
| RT @maiheurem: Very excited to share that the project I worked on during my internship @DeepMind got awarded a Distinguished Paper Award @R… | 2021-02-16 15:11:42 |
| RT @mer__edith: After firing @timnitGebru + @RealAbril + nurturing a racist and toxic culture, Google has the paternalistic AUDACITY to imp… | 2021-02-16 15:09:24 |
| RT @fchollet: The quickest way to lose the ability to do something is to stop believing you can do it. | 2021-02-16 15:06:02 |
| RT @maxplanckpress: Are you ready for Feb. 18? Here’s everything you need for #CountdownToMars @NASAPersevere Toolkit: https://t.co/Y1r… | 2021-02-16 15:04:15 |
| After firing @timnitGebru + @RealAbril + nurturing a racist and toxic culture, Google has the paternalistic AUDACITY to imply | 2021-02-16 15:02:19 |
| RT @giannis_daras: New paper: | 2021-02-16 15:00:57 |
| RT @togelius: The Piaggio P.180, beloved by deep learning enthusiasts because it has backprop https://t.co/jWsdDEDI8Z | 2021-02-16 14:57:51 |
| RT @FadaduSudeep: @ParikPatelCFA Highly recommend “rebooting AI” by @GaryMarcus and Ernest Davis. It reflects the current state of the AI a… | 2021-02-16 14:57:05 |
| Do what now? | 2021-02-16 14:56:53 |
| RT @paperswithcode: Introducing...Papers with Libraries! We are now indexing ML libraries from the community. We start with the amazing T… | 2021-02-16 14:55:39 |
| RT @page_gould: When you are using multilevel models to analyze data, you may come across a warning, “The Hessian Matrix is not Positive de… | 2021-02-16 14:50:45 |
| RT @techwontsaveus: Amsterdam has ruled Deliveroo drivers are employees. It will be interesting to see where things go after this decision.… | 2021-02-16 14:36:23 |
| Danielle reminding us (again) that we can and should do better at elevating awareness of the plight of the trans community in academia. Especially because some academics *actively promote* transphobia. https://t.co/njdmfAPoPN | 2021-02-16 14:34:36 |
| RT @emnlpmeeting: The call for demo papers at EMNLP 2021 is out! https://t.co/o4gIhx4ett | 2021-02-16 14:10:28 |
| @thomasforth I guess you could have said that your tweet was | 2021-02-16 13:53:09 |
| RT @meganakpeters: this beautiful, massive description of the content from @neuromatch Academy in summer 2020 is now up on OSF! special tha… | 2021-02-16 12:44:57 |
| RT @wiebketous: Just rewatched @shakir_za's NeurIPS 2020 presentation on surveys and writing in ML. Thank you for emphasising and expressin… | 2021-02-16 12:41:54 |
| RT @onucharlesc: We have internship positions available at @ubenwa_ai for people interested in using machine learning to tackle infant heal… | 2021-02-16 12:41:24 |
| RT @maiheurem: Very excited to share that the project I worked on during my internship @DeepMind got awarded a Distinguished Paper Award @R… | 2021-02-16 12:01:47 |
| RT @MogensAgerbo: | 2021-02-16 11:45:04 |
| RT @maiheurem: Very excited to share that the project I worked on during my internship @DeepMind got awarded a Distinguished Paper Award @R… | 2021-02-16 10:08:15 |
| RT @JMateosGarcia: The principle of requisite variety suggests that ML systems can't grasp social realities that are relational, complex an… | 2021-02-16 09:49:38 |
| RT @sarahookr: Yesterday, I ended up in a debate where the position was | 2021-02-16 08:30:05 |
| @mmitchell_ai @timnitGebru Sending endless energy & You're both my heroes. Thank you for fighting systems. It feels like the needle doesn't move, but I promise it's moved in strides | 2021-02-16 08:04:27 |
| RT @riggaroo: We are still hiring for a few open Android positions at @Over. The work is challenging and fun and I'd love it if you conside… | 2021-02-16 07:57:05 |
| RT @jackclarkSF: So when Google inevitably has to disclose more details about JFT and tries to make a big song and dance about its work on… | 2021-02-16 07:52:37 |
| RT @jackclarkSF: two of the world experts in documenting models and their underlying datasets work(ed) on Google's Ethical AI team - @timni… | 2021-02-16 07:52:30 |
| RT @jackclarkSF: It's ironic to me that more and more of Google's papers reference JFT, a secret in-house image dataset. JFT is going to be… | 2021-02-16 07:52:27 |
| RT @rcalo: It's been decades, but it's time to revisit space law's accounting of the public interest. This company and others are putting h… | 2021-02-16 07:33:23 |
| RT @ak92501: TransGAN: Two Transformers Can Make One Strong GAN pdf: https://t.co/EUTkjaEvF4 abs: https://t.co/vRTwLMNDPm github: https://t… | 2021-02-16 07:31:55 |
| RT @Grady_Booch: @unimelb @CAIDE_UoM My litmus test: an AI is a system that reasons and that learns. If one or both of these elements is m… | 2021-02-16 06:52:57 |
| @unimelb @CAIDE_UoM My litmus test: an AI is a system that reasons and that learns. If one or both of these elements is missing, it’s not an AI. | 2021-02-16 06:52:47 |
| @TBBake @tyrell_turing @KordingLab @PsychScientists @andpru @neuro_data @jpillowtime @neurograce Can I ask that if you choose to police our language you give some consideration to what definitions are useful to us, on top of philosophical considerations? Definitions that can never be disproven (like is it causal for some unspecified behaviour) are not useful. | 2021-02-16 06:39:57 |
| @manuelbaltieri @PsychScientists @andpru @neuro_data @KordingLab @jpillowtime @neurograce Q is some neurons. P is the pattern of their activities. R is still the real house. There is no S required in either example. | 2021-02-16 06:29:23 |
| RT @CarlDoersch: Just in time for NeurIPS, we've released code for CrossTransformers https://t.co/hIyH5jDjMl into the main Meta-Dataset rep… | 2021-02-16 05:14:59 |
| RT @KaliTessera: Excited to present my first preprint - | 2021-02-16 05:14:48 |
| RT @gslaller: @ilyasut Do I smell gpt4? | 2021-02-16 05:12:29 |
| RT @ilyasut: It helps to dream about ML | 2021-02-16 05:11:57 |
| RT @dancherp: After working on programming languages (PL) for machine learning (ML) for a few years: I’m super excited to switch gears to… | 2021-02-16 05:10:37 |
| RT @SpaceX: Deployment of 60 Starlink satellites confirmed https://t.co/apZ7oTOvNk | 2021-02-16 05:10:20 |
| And instead of following that clearly articulated path, @Google is promising to | 2021-02-16 05:04:16 |
| Meanwhile, all the way back in early December, @GoogleWalkout laid out the path that @Google should have taken, to make things right. https://t.co/wDiZsPBa6L | 2021-02-16 05:01:32 |
| This is awful -- awful for Meg, awful for Timnit, awful for their team, and awful for all of us. Meg and Timnit and their team should be putting their time and energy into their research. We're all missing out. https://t.co/KTsB6mSh6H | 2021-02-16 05:00:12 |
| RT @alexhanna: But have you told the people about the benefits of large language models???? https://t.co/Ru46pWyRBx | 2021-02-16 04:57:48 |
| @TBBake @KordingLab @jpillowtime @neuro_data @behrenstimb @neurograce Sure, there’s a lot in philosophy that won’t be productive at all for doing good science. Good science doesn’t need to have the same definitions as philosophy or adopt every history of verbal confusion wars. Good interdisciplinary work looks at meaningful and productive bridges. | 2021-02-16 04:57:40 |
| RT @huggingface: We've been getting lots of questions on how to contribute models toTransformers. Recently we started to publish model-s… | 2021-02-16 04:38:45 |
| RT @Miles_Brundage: Super excited to share a report I’ve been working on with a bunch of colleagues from different orgs since last year. Bu… | 2021-02-16 04:37:52 |
| This is a fascinating extrapolative model: https://t.co/9Qe9Y7Uxiv https://t.co/4KX7EX9P11 | 2021-02-16 04:10:34 |
| Awesome!! :) I will unfortunately be in the middle of real work at 3pm PST. Which points to the biggest drawbacks of Clubhouse atm - no option of recording when it makes sense, and ofc iPhone / invite issues. Love the ease of use and interactivity the platform affords though! https://t.co/XUHSwmGUKO | 2021-02-16 03:31:07 |
| RT @roydanroy: Nice all-student paper out of Toronto. Impressive work, IMO. | 2021-02-16 02:37:00 |
| RT @wewatchwatchers: aliyyah abdur-rahman and me, talking Black Privacy in the new issue of The Black Scholar https://t.co/7hmUD7ZGUq | 2021-02-16 02:20:21 |
| RT @negar_rz: 11 weeks after my first manager got fired and 5 weeks after my second manager is locked out of work! | 2021-02-16 02:15:12 |
| RT @negar_rz: 11 weeks after my first manager got fired and 5 weeks after my second manager is locked out of work! | 2021-02-16 02:11:59 |
| RT @rajiinio: This @NewYorker article heavily features our #NBIAIR workshop @NeurIPSConf & | 2021-02-16 02:09:36 |
| @neuro_data @TBBake @KordingLab @jpillowtime @behrenstimb @neurograce @_MiguelHernan As others have pointed out, the notion of representation itself does not require nor does it imply a causal link to behavior. | 2021-02-16 02:04:57 |
| RT @Montreal_AI: A Too-Good-to-be-True Prior to Reduce Shortcut Reliance Dagaev et al.: https://t.co/qjNfLrA6WF #ArtificialIntelligence #… | 2021-02-16 02:02:49 |
| @behrenstimb @KordingLab @PsychScientists @andpru @neuro_data @jpillowtime @neurograce Now Tim, philosophers are literally paid to make problems out of non-problems by engaging in confusing language games... they're only doing their jobs, be nice. | 2021-02-16 01:45:04 |
| @HamelHusain @kjw_chiu @pvierth Generally you can just use `asyncio` instead of `select` directly. IIRC `asyncio` behind the scenes uses epoll/kqueue/IOCP/etc so it scales well. For extra performance you can replace the event loop with libuv https://t.co/pZNc6C3S1Y | 2021-02-16 01:43:52 |
| RT @alexhanna: Not sure how Google really expects to rebuild trust in folks working in ethics when we're still dealing with this. Mind you… | 2021-02-16 01:30:00 |
| The idea that more information and connections can, perhaps paradoxically, lead to more polarization (or balkanization) is something @InfoEcon and I explored in our earlier paper: https://t.co/QFAHIZW7hY | 2021-02-16 01:21:31 |
| @TobyWalsh Fortunately, there is a growing research community looking at ways to vastly reduce the cost of training these models. I predict brute force will give way to much more efficient algorithms, implementations, and hardware in the next few years. | 2021-02-16 01:15:19 |
| This new NBER paper helps explain polarization: | 2021-02-16 00:57:17 |
| RT @FrankPasquale: Harassment of Black feminists became a blueprint for what would eventually become Gamergate, “which became a blueprint f… | 2021-02-16 00:47:38 |
| RT @rcalo: Many schools that confer PhDs require that a member of the committee come from outside the discipline. Serving on these committe… | 2021-02-16 00:36:36 |
| Dev is what you do on the journey to Ops. | 2021-02-16 00:31:55 |
| RT @mmitchell_ai: My career is possible in part because of an intensely honed ability to see and work with patterns. This is arguably why I… | 2021-02-16 00:20:09 |
| RT @TopicalGamer: Tokyo now seems to have craft beer vending machines. First time I've seen one https://t.co/bxpyRne5xE | 2021-02-15 02:55:44 |
| @conor_muldoon @JojoFreebanks @DavidDeutschOxf If. And yet, human intelligence exists in this universe, with its laws of physics. | 2021-02-13 01:13:14 |
| Deadline extended for the Human-Centered XAI workshop https://t.co/yVe8KDTtD7 | 2021-02-13 01:10:42 |
| RT @jake_fisher: *taps mic* hello to my cognitive neuroscience friends https://t.co/KVH0bUIorl | 2021-02-13 01:03:17 |
| They have this paper: https://t.co/lIaCa6armZ as proof of work related to environmental impact. I haven’t even read the part where they talk about feedback given to me cause other people are already seething from this letter and I know I will take a while to recover. | 2021-02-13 00:57:24 |
| I have so much to say about this letter each glimpse I take. They (eg) have some extremely toxic men who they apparently couldn’t fire cited in there. @sundarpichai @JeffDean this your plan to garner employee trust? Gals-lighters in chief they have absolutely no shame. https://t.co/YvhywkSYh2 | 2021-02-13 00:54:46 |
| RT @radekosmulski: THREAD: Sometimes code can seem impenetrable. But there are various ways we can make our life easier. A common pattern… | 2021-02-13 00:46:36 |
| RT @sohamde_: Releasing NFNets: SOTA on ImageNet. Without normalization layers! https://t.co/eUPSNle0r2 Code: https://t.co/YuwULNqpTR Thi… | 2021-02-13 00:39:50 |
| I bet they were gonna add model cards & Still haven’t gotten myself to read the whole letter cause I know it’ll make me sick. https://t.co/3ttPKVnTnz | 2021-02-13 00:06:43 |
| RT @mrfrank5790: Modeling the new hoodie from @DigEconLab !! https://t.co/KevYg9pvBY | 2021-02-12 23:34:45 |
| RT @labortechgroup: We demand #transparency to the broader public, including @Google users and our colleagues in the academic community, ab… | 2021-02-12 23:34:06 |
| RT @dgorganize: 7 weeks of voting amid the ruthless union-busting of Amazon. That’s a long time. Hoping workers can find some restorative m… | 2021-02-12 23:33:36 |
| RT @Mehtabkn: CFP for 2021 CVPR workshop on “Beyond Fairness: Towards a Just, Equitable, and Accountable Computer Vision” organized by @cep… | 2021-02-12 23:29:36 |
| RT @Montreal_AI: https://t.co/VaMGbOjQ11 SPACE | 2021-02-12 23:26:13 |
| RT @Montreal_AI: https://t.co/VaMGbOjQ11 SPACE | 2021-02-12 23:26:01 |
| @moonriseTK @maroftmec @philiprosedale @Tesla you are comparing the reality of bitcoin's energy cost with the potential of solar. apples and oranges ATM | 2021-02-12 23:24:15 |
| RT @labortechgroup: #Google’s retaliation against @timnitGebru, & | 2021-02-12 23:23:21 |
| RT @Gizmodo: Minneapolis Bans City Cops From Using Face Recognition Tech https://t.co/i4MriFebmr https://t.co/Kduif9wGFk | 2021-02-12 23:19:19 |
| @moonriseTK @maroftmec @philiprosedale @Tesla again, bitcoin might be mined from renewable sources, but the energy spent mining, once spend no matter the source, is lost | 2021-02-12 23:14:51 |
| @moonriseTK @maroftmec @philiprosedale @Tesla you make the mistake that most bitcoin mining comes from solar power China and North Korea would disagree with you.... | 2021-02-12 23:04:30 |
| RT @schemeprincess: And @MSFTResearch Systems Group is hiring! | 2021-02-12 22:57:23 |
| RT @Montreal_AI: Deep RL with OpenAI Gym 101 • Session 0 · Getting Started with OpenAI Gym 101 • Session 1 · Evolution Strategies 101 • Se… | 2021-02-12 22:56:44 |
| @onaclov2000 @terrible_coder Some say the number of parameters is overkill. | 2021-02-12 22:56:32 |
| RT @luislamb: ‘The Debate of the Next Decade’ — AI Debate 2 Explores AGI and AI Ethics - Synced #AIDebate #AIDebate2 #MontrealAI @ceobillio… | 2021-02-12 22:55:34 |
| @graemedmoffat Sure, from an economics perspective, I totally agree. That's why we need more *real* home-grown champs! But, for the individual researcher, the environment and support here for research itself can be great. | 2021-02-12 22:53:41 |
| RT @sohamde_: Releasing NFNets: SOTA on ImageNet. Without normalization layers! https://t.co/eUPSNle0r2 Code: https://t.co/YuwULNqpTR Thi… | 2021-02-12 22:44:03 |
| RT @FlorianGD: What happens to @timnitGebru is so shameful https://t.co/WXh7ep2j6B | 2021-02-12 22:14:28 |
| @ccaballeroh10 yeah CL is its whole separate (and rather dormant...) field. but still the main nlp journal used to be CL, the association is ACL, the conferences are xCL, ... | 2021-02-12 22:09:18 |
| RT @rcalo: Love it when scholars like @robseamans actually roll up their sleeves and get into the data. Check out what Rob and colleagues f… | 2021-02-12 21:59:34 |
| @graemedmoffat But, *only* because we do have a lot of big tech companies with facilities here. If we didn't, then it would indeed be a pretty bleak landscape, cause the gov funding and home-grown corporate R& | 2021-02-12 21:37:34 |
| @graemedmoffat Oh, indeed, but I suspect that the Canadian expenditures are also listed in the totals for those companies, no? So, those numbers are not mutually exclusive, and makes it harder to draw conclusions from this data. Frankly, I think Canada is a great place to do tech research. | 2021-02-12 21:36:43 |
| Can you imagine what their “investigation” is gonna come out and say? And they had the nerve to contact me and ask for my “perspective” https://t.co/LkAfp66stL | 2021-02-12 21:36:26 |
| It’s going to take me some energy to actually read this letter because they are shameless. I mean it was already obvious I know. But each new thing is just incredible. And tomorrow Jeff Dean is gonna turn around & | 2021-02-12 21:35:09 |
| RT @mmitchell_ai: Does anyone know what the Google letter is talking about, where it says | 2021-02-12 21:29:43 |
| RT @Kensan42: They are but the platforms just suspend these people and let them back on after a while. https://t.co/WOdPnCOfzb | 2021-02-12 21:29:15 |
| Does anyone know what the Google letter is talking about, where it says | 2021-02-12 21:29:02 |
| RT @peard33: The investigation appears to be ongoing. But spokesman for @RonWyden says | 2021-02-12 21:25:42 |
| RT @marypcbuk: firing researchers for social responsibility? https://t.co/GPgKPDGqwh | 2021-02-12 21:24:50 |
| I stand with StarGirl | 2021-02-12 21:13:25 |
| RT @labortechgroup: Given the potential for bias to creep into AI systems (https://t.co/SXxoTmHcBG), tech companies have a societal respons… | 2021-02-12 21:13:09 |
| RT @labortechgroup: To uphold the future of #AI ethics, tech companies should open the company to outside researchers and support underfund… | 2021-02-12 21:13:00 |
| RT @labortechgroup: We call on @Google Research to strengthen its commitment to research integrity and to unequivocally commit to supportin… | 2021-02-12 21:12:55 |
| RT @labortechgroup: We, the undersigned, stand in solidarity with Dr. @timnitGebru, who was terminated from her position as Staff Research… | 2021-02-12 21:12:49 |
| So exciting to see that people who buy #GirlDecoded buy @BarackObama’s #memoir Love our common themes of #empathy, #hope, #resilience & @penguinrandom #author #HumanFirst #IAmGirlDecoded #EmotionAI #technology https://t.co/MvV3bNnTTp | 2021-02-12 20:37:02 |
| RT @TCIAMN: We did it, Minneapolis. The City Council just voted to ban facial recognition in our city — a huge victory for privacy and for… | 2021-02-12 20:28:59 |
| RT @suryamattu: Good example of why we invest the time in building tools that allow for persistent monitoring https://t.co/9QHn9fVxnG | 2021-02-12 20:26:49 |
| RT @awjuliani: I enjoyed reading | 2021-02-12 20:24:06 |
| RT @catherine_regis: My new paper with @lyse_langlois is out Analyzing the Contribution of Ethical Charters for the Future of AI Gover… | 2021-02-12 20:23:11 |
| @graemedmoffat One caveat: many of these US tech companies have research offices in Canada. | 2021-02-12 20:13:51 |
| @boazbaraktcs @srush_nlp @davidweichiang I shudder when reading pytorch code to see integers for axes. We should at least use named constants, but even better would be names that are checked by the interpreter | 2021-02-12 20:12:26 |
| @dkaushik96 @mataneyal1 @asaf_amr @hilleltt i am aware of many of these lines of work, but what are the relations you see to the ones by zori? | 2021-02-12 20:06:07 |
| @RecklessCoding In this instance yes. Filters only catch the most obvious, extreme offensive language | 2021-02-12 20:02:48 |
| RT @triciawang: Must read article about big tech's role in creating a fascist and racist system from @nantarsya @mer__edith takeaway: we… | 2021-02-12 20:02:00 |
| Microsoft is building a 90 acre campus to the west of Georgia Tech campus https://t.co/iKAwoH8zDI Will dedicate 22 acres of it to community use. | 2021-02-12 20:00:37 |
| They're using words.. sometimes sentences.. it's the real deal! https://t.co/9s6Uydgyk7 | 2021-02-12 19:59:30 |
| RT @awjuliani: I enjoyed reading | 2021-02-12 19:52:36 |
| @neuralreckoning @Timothy0Leary I would say borderline inappropriate (unless there was a promise of payment) rather than unethical. If they want to farm their program requirement checking out to other people, they should pay for it. | 2021-02-12 19:26:45 |
| RT @annierpalmer: New: I spoke to Amazon delivery drivers about an app they use called Mentor, which is framed as a driver safety tool, but… | 2021-02-12 19:23:50 |
| When you help one another recover from exceptions. They never let your stack overflow. They are open to mutual refactoring. When the halting problem is always dismissed as a theoretical problem. #ValentinesDayForProgrammers https://t.co/S9gy5ysLb1 | 2021-02-12 19:20:20 |
| RT @ak92501: High-Performance Large-Scale Image Recognition Without Normalization pdf: https://t.co/THe2NfRI1K abs: https://t.co/Z68FevANZP… | 2021-02-12 19:07:36 |
| RT @maier_ak: I guess this is another point for @GaryMarcus... https://t.co/GzYFQjU2nx | 2021-02-12 19:05:49 |
| RT @RobinSandfort: The future is now. Camera trapping, Machine Learning and LoRa connectivity on the edge! Exiting times. #tech4wildlife @a… | 2021-02-12 18:58:09 |
| RT @MontrealRobots: The next speaker in the Robot Learning Seminar Series is @SylviaLHerbert from @UCSDJacobs Tune in at 4 pm ET today (1… | 2021-02-12 18:57:09 |
| RT @DataScienceNIG: Welcome, TensorFlow 3D from @GoogleAI, an open-source framework of training and evaluation pipelines for state-of-the-a… | 2021-02-12 18:57:00 |
| @terrible_coder I thought GPT-3 was supposed to be zero shot! | 2021-02-12 18:25:51 |
| RT @stephenharred: Can confirm: “The truth is at Google you ascend by bullying your reports.” https://t.co/wTurAQNdT4 | 2021-02-12 18:04:37 |
| RT @jftaveira1993: Men in the tech world are 'harassing' and 'stalking' @timnitGebru following her ouster from #Google. https://t.co/T8wMXu… | 2021-02-12 17:57:46 |
| Dylan was my direct report and they had the audacity to say something about me as a manager in their email terminating me. The truth is at Google you ascend by bullying your reports. https://t.co/z61ZWsukti | 2021-02-12 17:57:34 |
| @dylnbkr Awww thank you friend. I'm sorry about everything thats happening right now its a nightmare. Why they consistently choose to be so horrible who knows. | 2021-02-12 17:56:22 |
| Excited that today at 4 PM ET (in 3 hrs) @caswellcaswell will talk abt spatial navigation! Join us: https://t.co/xW87BZ1soe Neurobiological successor features for spatial navigation https://t.co/fMlVk7ECE4 Interpreting Wide-Band Neural Activity Using CNN https://t.co/p2o2izo71b https://t.co/chUyMQGvLd | 2021-02-12 17:53:01 |
| Thank you Ifeoma, @SareetaAmrute and @gleemie for writing this. https://t.co/B5Sgh2GtCq | 2021-02-12 17:41:49 |
| RT @iajunwa: In this piece for @nature, my co-authors and I observe the intersection of labor rights and racial equity in the exit of @tim… | 2021-02-12 17:37:04 |
| RT @databoydg: QA redacted: but there was a really interesting question about how we can prevent our communities from being | 2021-02-12 17:36:45 |
| HA | 2021-02-12 17:36:21 |
| RT @quaesita: A free #AI course that puts the fun in fundamentals! Sign up here: https://t.co/wWNzyJ2NvO - the first 90min arrives in a fe… | 2021-02-12 17:22:42 |
| RT @Mila_Quebec: AI for the Planet - February 16th - 2 pm CET A global virtual conference organized by @Startup_inside_ in collaboratio… | 2021-02-12 17:22:33 |
| RT @PartnershipAI: Today, we welcome new Partners from civil society, academia, and industry to build a healthy information ecosystem. Thes… | 2021-02-12 17:22:07 |
| RT @DeepMind: Introducing NFNets, a family of image classification models that are: *SOTA on ImageNet (86.5% top-1 w/o extra data) *Up to… | 2021-02-12 17:22:01 |
| RT @neverendingrl: We invite you to submit papers (up to 6 pages, excluding references and appendix) in the @iclr_conf format. Submission… | 2021-02-12 17:21:52 |
| RT @clured: Excited to release a big update to the @opensyllabus Galaxy visualization today, which shows node2vec -> | 2021-02-12 17:21:44 |
| What makes a SasS company valuable isn't really its software (which could be cloned), it's its contracts, its reputation, its customer funnel... buying a SasS company | 2021-02-12 17:19:56 |
| RT @maier_ak: I guess this is another point for @GaryMarcus... https://t.co/GzYFQjU2nx | 2021-02-12 17:18:40 |
| RT @BU_Computing: Sabelo Mhlambi talks about the recent marginalization of @timnitGebru & | 2021-02-12 16:51:59 |
| @ampanmdagaba @iandanforth The latter @iandanforth is something I am looking into now in my research. | 2021-02-12 16:49:01 |
| @ampanmdagaba @iandanforth A lot of attempts at training language models to reduce offensive language don’t work terribly well because they only learn surface features. Same reason why the generate offensive language in the first place… their model isn’t very sophisticated about implications | 2021-02-12 16:48:29 |
| @nlpnoah *Human* Language Technologies. | 2021-02-12 16:47:14 |
| @ampanmdagaba @iandanforth The former is possible https://t.co/QEkDYDIXew *if* one has a model of acceptable speech and the language model can generalize from it... | 2021-02-12 16:46:57 |
| @_inesmontani @tiangolo i find that i can do it, and it sometimes even help - but must be specific kinds of coding, and music must be rhythmic, vocal-less, and repetitive. | 2021-02-12 16:46:22 |
| @tracyfrey @_d3v4n5h @mmitchell_ai @JeffDean It’s not just your silence. People currently terrorized know when someone is putting something in the line & Anyways right now I should take care of myself & Thanks. | 2021-02-12 16:35:55 |
| A bunch of new partners with the Partnership on AI. https://t.co/oX99UqaWOq | 2021-02-12 16:23:26 |
| 1. Language models don’t know “why” they say the things they say. 2. Language models don’t know why they shouldn’t say the things that they say. | 2021-02-12 16:18:57 |
| RT @gleemie: Is your city buying back light poles from utilities? Early warning that smart surveillance and monetized data is coming for yo… | 2021-02-12 16:06:16 |
| @VenkRamaswamy @TPVogels I just shared the OBS virtual camera | 2021-02-12 16:01:18 |
| In an hour, at 12pm ET Feb 12, join me! https://t.co/KXrKGQrpcR | 2021-02-12 15:49:49 |
| Good thing no one is seriously considering commercializing GPT-3 https://t.co/eOG8g2yw0J | 2021-02-12 15:38:17 |
| RT @jilliancyork: Actually, @HelleThorning_S, content moderation is almost always done | 2021-02-12 15:08:35 |
| @QueerZoomer @teamretrorabbit @phlippieb I think you're the only person I know who fits in the intersection of ML twitter and shoegaze | 2021-02-12 15:08:08 |
| @Atsiko yeah its a subset. (and | 2021-02-12 14:57:19 |
| RT @smfleming: My book on the science of self-awareness and metacognition, | 2021-02-12 14:52:59 |
| RT @hugospiers: Excellent new review from @clairegillanTCD & Smartphones and the Neuroscience of Mental Health https://t.co… | 2021-02-12 14:52:42 |
| RT @TrevorBack: After 9 incredible years @DeepMind it's time for me to say farewell and thanks! It's been a privilege to have been part of… | 2021-02-12 14:52:38 |
| RT @ajmooch: Our most recent work on training Normalizer-Free nets! We focus on developing performant architectures which train fast, and… | 2021-02-12 14:51:13 |
| @jenlhoy @ZeroNoiseLab Hahaa IMHO we can’t say with any certainty that mice don’t have consciousness. | 2021-02-12 14:43:25 |
| @jenlhoy @ZeroNoiseLab Not crazy at all! I’ve heard folks say over the years that the use of consciousness may be for those moments when things go wrong and you need all your brain to agree to adjust perception and/or action. Fits global workspace ideas of consciousness as broadcasting too. | 2021-02-12 14:36:47 |
| @djnavarro @IrisVanRooij Tbh I think the focus of topics, types of interdisciplinary work, and conventions are different across the two. :) | 2021-02-12 14:33:31 |
| @alyndanoel @neurobongo Agreed. I was quite taken by surprise at the certainty of some of the assertions. And I felt that prediction error consciously :) | 2021-02-12 14:23:01 |
| The talented Karin Brisker @BriskerKarin graduated her masters (on sequence modelling for cybersecurity and biology) during covid a few months ago, and today I received this lovely piece of NLP tech and croc-figures. (I would say | 2021-02-12 14:19:26 |
| (small)Realization: NLP is an OK name, much more fitting than CL (Computational Linguistics), but really, HLT (Human Language Technologies) is a so much more fitting name to what we do. Pitty it didn't catch. Do you think its too late to change? | 2021-02-12 14:09:30 |
| RT @atg_abhishek: So sometimes working on #responsibleAI can feel overwhelming! There are so many things to do and it is not clear on how a… | 2021-02-12 14:02:06 |
| @ptullochott @MSF_EastAfrica @ZekuZelalem Our student @bonjora has become a foremost expert in this type of work. We can hopefully publish her work soon. | 2021-02-08 00:39:36 |
| @geomblog @Aaroth For academic seminars, I love the proliferation of open, international seminars, and there is no reason to travel to do those. 3/ | 2021-02-08 00:26:14 |
| @geomblog @Aaroth I love being able to watch a 3-5minute short talk about each poster at NeurIPS and ICML. A new publication model should include the video as well as the paper and the Jupyter notebook (and maybe a machine-checkable proof?) 2/ | 2021-02-08 00:25:19 |
| @geomblog @Aaroth I think we need physical conferences to build and maintain the social structure of scientific fields. But maybe we can design them to optimize for that rather than for also being refereed publications. 1/ | 2021-02-08 00:23:27 |
| Some of my code is a tragedy But a lot of my code generates significant laughter. https://t.co/x564Olj1s5 | 2021-02-08 00:19:10 |
| RT @neilzegh: I will present LEAF, a learnable frontend for audio classification, at ICLR 2021. * Learns filtering, pooling, compression, n… | 2021-02-07 23:37:20 |
| RT @Jeande_d: A while ago, I completed Deep Learning Specialization taught by @AndrewYNg and @DeepLearningAI_. The entire specialization w… | 2021-02-07 23:33:25 |
| RT @karpathy: Everyone is so obsessed with accelerating neural nets, so as a fun side project I've been building this breadboard 8bit neura… | 2021-02-07 23:32:26 |
| RT @KordingLab: @neuro_data @danilobzdok @AndrewYNg @Montreal_AI Yes. Totally. But carpentry can readily be learned. artistry is hard. | 2021-02-07 23:31:13 |
| RT @neuro_data: @KordingLab @danilobzdok @AndrewYNg @Montreal_AI Data science needs both carpentry and artistry. Like everything, success d… | 2021-02-07 23:31:10 |
| RT @danilobzdok: @AndrewYNg: the type of #AI #job that is most difficult to find right now, the person how can look at a given business c… | 2021-02-07 23:31:06 |
| WATCH: Four Israeli scientists unexpectedly win Oscar for contribution to film https://t.co/o9yjxwnXzz | 2021-02-07 23:15:36 |
| RT @profcikara: With all due respect, I’m aware of all of this. I have organized several conferences, preconferences, and seminars. But the… | 2021-02-07 22:58:45 |
| @jasonbaldridge @ai_habitat [provided that I don't want to spend the next 10 years in data collection and annotation projects] | 2021-02-07 22:51:50 |
| @jimcoan @profcikara @jayvanbavel I’m still owed $2000 for conference fees from columbia | 2021-02-07 22:47:54 |
| @AravSrinivas Agree, I like the mess of colors! Maybe a few dozen hours so far but spread out over months because FSD. I was able to get away with a voltmeter so far. I know debugging would be a nightmare so I go very slow and triple check everything, which has worked so far. | 2021-02-07 22:38:41 |
| @jasonbaldridge @ai_habitat so, these were on the first half of the talk.. but then you went on to discuss a bunch of other linguistic phenomena (say, tenses, metaphorical usage of concrete as abstract, etc). how does one go about modeling these in a meaningful way (...in an interactive env?) | 2021-02-07 22:33:11 |
| RT @Greene_DM: Really appreciate this genealogy of ImageNet from @alexhanna and friends. This is what we should be doing. | 2021-02-07 22:15:35 |