Vlad Mnih

It's been 300 days since the start of Russia's full scale invasion of Ukraine. Currently millions of Ukrainians go days without electricity, heat and water due to Russia's attacks. While Ukraine has liberated much of its territory it will need continued support to win and rebuild

SIMA 2 is our most capable AI agent for virtual 3D worlds. 👾🌐 Powered by Gemini, it goes beyond following basic instructions to think, understand, and take actions in interactive environments – meaning you can talk to it through text, voice, or even images. Here’s how 🧵

Today I’m launching @reflection_ai with my friend and co-founder @real_ioannis. Our team pioneered major advances in RL and LLMs, including AlphaGo and Gemini. At Reflection, we're building superintelligent autonomous systems. Starting with autonomous coding.

I am happy to advertise a position for a student researcher under the Google DeepMind program #student-researcher-program" target="_blank" rel="nofollow noopener">deepmind.google/about/careers/… , for a project with @TZahavy , myself, and others, at the intersection of generative modelling, creativity, and planning. Strong programming skills required.

@dmarthal @hugo_larochelle Yes, international applicants are eligible.

Are you interested in building generalist agents? Our team is hiring for the roles of Research Engineer and Research Scientist in Toronto. Apply at deepmind.com/careers/jobs?l…

@Chad_R_Stewart @TechIsHiring Yes, feel free to retweet. Thank you!

@VladMnih Hey there! I run a space @TechIsHiring in an attempt to consolidate tech job posting tweets and searching for tech job tweets into one place. Do you mind if I quote tweet your post with the TechIsHiring hashtag?

Our team is hiring for both RS and RE roles! Research focus is building generalist agents. At ICML this week, ping me if you're interested in chatting.

One of the most direct and transparent ways to help is through United24 @U24_gov_ua, which was launched by the President of Ukraine as the main venue for collecting donations for defense, medical aid and rebuilding. u24.gov.ua

It's been 300 days since the start of Russia's full scale invasion of Ukraine. Currently millions of Ukrainians go days without electricity, heat and water due to Russia's attacks. While Ukraine has liberated much of its territory it will need continued support to win and rebuild

Transformers have shown remarkable capabilities - but can they improve themselves autonomously from trial and error? New research shows how a single transformer can explore and solve tasks using this method without ever updating its parameters ➡️ dpmd.ai/3sv7Oh1

Really excited to share our recent work on in-context reinforcement learning with transformers - arxiv.org/abs/2210.14215

Congratulations to @tejasdkulkarni and the @CSM_ai team! CommonSim-1's Real2Sim capabilities look really impressive.

Kyiv-Mohyla Academy, one of the best Ukrainian Universities, is collecting a scholarship fund for students critically affected by the war (lost relatives/home/were displaced). If you want to support 🇺🇦 education, consider donating to them (PayPal/SWIFT): kma-scholarships.org/en/

We have two research engineer positions in the vision group at DeepMind. Apply at boards.greenhouse.io/deepmind/jobs/…

Gato🐈a scalable generalist agent that uses a single transformer with exactly the same weights to play Atari, follow text instructions, caption images, chat with people, control a real robot arm, and more: dpmd.ai/Gato Paper: dpmd.ai/Gato-paper 1/

If you’re interested in unsupervised skill learning, exploration, variational infomax, and/or ensemble-based uncertainty estimation, come chat with us at our #ICLR2022 poster today! (Starting in just over an hour…) iclr.cc/virtual/2022/p… @Zergylord @katebaumli @dwf @VladMnih

Are you a foreigner wanting to help Ukraine from abroad? Then this list is for you. All the verified ways to help Ukraine and Ukrainians euromaidanpress.com/2022/02/27/ver…

I'm looking to hire Research Scientists to work on all things deep reinforcement learning with me on the Discovery team @deepmind. If you are interested in exploration, options, meta-RL, generalization, etc. apply here and mention me in your application: boards.greenhouse.io/deepmind/jobs/…

are you a student from Ukraine facing program interruption, apply to @UofTCompSci Summer Program for Students from Ukraine. (@VectorInst, @UofT) All expenses covered and quick turnaround. 📅 Apply by Apr 8, a decision by Apr 13. web.cs.toronto.edu/summer-program… 🙏spread the word🙏

Unsupervised skill learning allows RL agents to learn useful behaviors in the absence of task-specific rewards. But beneath its promising facade lies a dark secret… a challenging exploration problem! Check out our #iclr2022 Spotlight to learn more: arxiv.org/abs/2107.14226. 🧵⬇️