Infer2Control, our workshop on Probabilistic Reinforcement Learning and Structured Control #nips2018, had its deadline extended to this Fri, Oct 12. A limited number of NIPS tickets are reserved for accepted talks and posters. sites.google.com/view/infer2con…

Releasing Spinning Up in Deep RL, a new educational package to let anyone become a deep reinforcement learning practitioner: blog.openai.com/spinning-up-in… Next year, we'll teach workshops on Spinning Up at CHAI, OpenAI, and other venues.

My colleague Vincent Vanhoucke wrote up a very nice blog post on what it means to be a researcher, and some of the differences between research and engineering. medium.com/@vanhoucke/so-…

"Soft Actor-Critic Algorithms and Applications," Haarnoja, Zhou, and Hartikainen et al.: arxiv.org/abs/1812.05905

My thesis "Acquiring Diverse Robot Skills via Maximum Entropy Deep Reinforcement Learning" has been accepted! Below is a 17 second summary. Thanks to my committee Sergey Levine, Pieter Abbeel, Mark Mueller and all the collaborators! www2.eecs.berkeley.edu/Pubs/TechRpts/…

Great WIRED article about our research on real-world RL for robotic walking, with some nice quotes from Tuomas Haarnoja: wired.com/story/the-clev… (research w/ Tuomas Haarnoja, Aurick Zhou, George Tucker, Sehoon Ha, and Jie Tan)

Videos of our #Infer2Control workshop #neurips2018 are now available, featuring great talks by Ryan Adams Ryan Adams, Anne Collins Collins Lab, Leslie Kaelbling, Sergey Levine Sergey Levine, Shie Mannor, Emo Todorov, and many others! youtube.com/playlist?list=…

Big update to Stable-Baselines: v2.4.0 is out! Highlights: - New RL baseline: Soft Actor-Crtic (Tuomas Haarnoja et al.) with automatic entropy temperature - Customize policy at model creation - RL Baselines Zoo updated (now 80+ trained agents) Full changelog: github.com/hill-a/stable-…

Excited to share I have joined DeepMindAI in London as a Research Scientist!

L2 distance is a bad proxy for how long it takes to reach a goal. Can we learn dynamical distances? DDL can shape rewards and help learn from preferences. Below: robot learning w/ DDL from images w/o reward, only prefs. sites.google.com/view/skills-vi… w K Hartikainen, X Geng, Tuomas Haarnoja

CoRL registration has just opened! Looking forward to the only conference devoted to machine learning for robotics, this autumn in Osaka, Japan. #corl2019 robot-learning.org

A sneak peek to our latest work; more details to follow soon. Stay tuned.

We just released our work on robot soccer. I've been working on this for quite some time with my amazing colleagues at DeepMind. It's exciting how deep RL can produce such beautiful behaviors with low-cost robots. Full paper is available at arxiv.org/abs/2304.13653 Enjoy!

Very proud to be part of this team! And looking forward to seeing more robots being... well, less robotic

Yeah! ScienceRobotics cover! science.org/doi/10.1126/sc…

New research from Google DeepMind brings together soccer and robotics. Using reinforcement learning, robots display agile and reactive movements similar to a soccer player, no shin guards needed:) science.org/doi/10.1126/sc…

It was a great honor to have you visit our lab!

Can deep reinforcement learning enable autonomous #robots 🤖🦿 to play real-world soccer ⚽️? sites.google.com/view/vision-so… Thrilled to share our latest step towards learning multi-agent robot soccer purely with onboard computation and sensing. We're extending prior motion capture