Our new AI system learned speech recognition in English with *zero* speech to text training data: researchers just gave it lots of audio, and it figured out what the words were. But it goes way beyond that - it learned Swahili too!
Our paper on adversarial motion priors has been accepted into @siggraph 2021! Adversarial imitation learning is finally starting to work for complex simulated characters.
Paper + code: xbpeng.github.io/projects/AMP/
Thanks to my collaborators: Ze Ma, @pabbeel@svlevine@akanazawa
We also talked a bit about why I started studying AI :)
Episode resources page put together by twiml:
twimlai.com/reinforcement-…
Thanks for having me on @twimlai!!
5/5
A single state leaks information about the reward function. We can learn from it by simulating what might have happened in the past that led to that state (previously in small toy environments, now the scaled-up version in slightly less-toy environments :) @interact_ucb
How can RL agents explore *safely*? Conservative safety critics aim to provide this via a pessimistic safety critic that believes that anything you haven't done before is dangerous. At #ICLR2021 today, 5 pm PT, rm B5!
arxiv: arxiv.org/pdf/2010.14497…
ICLR: iclr.cc/virtual/2021/p…
To improve RL, we try to build better algorithms. But what if we ask how to set up environments to make RL (esp. lifelong RL w/ sparse rewards) easier? Check out Ecological RL, which was invited to the never-ending RL ws: sites.google.com/view/neverendi…#ICLR2021arxiv.org/abs/2006.12478
RL agents can learn "adversarial behaviors" that trigger pathologically irrational responses in *other* agents in their environment. Read more about these physically realistic attacks in a new BAIR blog post by Adam Gleave!
The paper on meta-world -- a new benchmark for meta-learning with 50 distinct robotic manipulation tasks -- is now out!
Code & documentation here: meta-world.github.io
Full paper: arxiv.org/abs/1910.10897
Want to learn about offline RL? Check out our new tutorial on offline RL, w/ Aviral Kumar, @georgejtucker, Justin Fu: arxiv.org/abs/2005.01643
Offline RL may enable RL algorithms to use large offline datasets, and thus make it applicable to a wide range of real-world problems.
580K Followers 169 FollowingSalesforce is the #1 Agentic CRM — bringing humans, agents, and platforms together to drive customer success. Tweet @AskSalesforce for help.
11K Followers 556 FollowingHead of AI @monacoGTM, @stanford teaching AI productivity for 32K+ devs https://t.co/HsYzrfFNPS, YC S24, @ConfettiAI (acq'd), ML @amazon @stanfordnlp
22K Followers 1K FollowingProfessor @ucsantabarbara. Head of Research @SimularAI. Director @ucsbcrml @UCSB_AI. Build the Science of Multimodal AI Agents. AI for Humanity in the long run.
6K Followers 81 FollowingSince 1987, the top international conference in the computer vision field. Co-sponsored by the IEEE and the Computer Vision Foundation.
31K Followers 93 FollowingFounded in 1979, AAAI is an international, nonprofit, scientific society devoted to promote research in, and responsible use of Artificial Intelligence.
251K Followers 2K FollowingThe world's leading publication for data science and artificial intelligence professionals.
Submit an Article ✍️ https://t.co/57pIMegK1o
318K Followers 288 FollowingKaggle is the largest global AI community of developers, researchers, and enthusiasts who compete, collaborate, and benchmark what's next in AI.
33K Followers 6 FollowingFounder and CEO of @insitro, Machine Learning pioneer, co-founder of Coursera, adjunct CS Professor at Stanford, avid traveler
450K Followers 6K FollowingChief Scientist, Google DeepMind & Google Research. Gemini Lead. Opinions stated here are my own, not those of Google. TensorFlow, MapReduce, Bigtable, ...
1K Followers 184 FollowingAsst. Professor of Chemical and Biomolecular Engineering @UCLA @UclaCBE | Inventing new materials and tools for energy sustainability | PhD '18 @StanfordMSE
23K Followers 476 FollowingAssociate Professor @UTCompSci | Director @NVIDIAAI Co-Leading GEAR | CS PhD @Stanford | Building generalist robot autonomy in the wild | Opinions are my own