🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

🚨 Reminder! Submissions for RL_Conference's Finding the Frame are due May 30 (AoE)! We're looking for bold ideas that rethink the foundations of RL: goals, values, rewards, formalisms, and beyond. 🧠Philosophy, theory, critique welcome! 🔗More details: sites.google.com/view/findingth…

🤔 Unsure whether your work fits the spirit of the workshop? ✨ Check out Micah Carroll ’s paper from last year’s workshop on rethinking fundamental assumptions on reward functions! ———— ⏳ Submit by May 30 (AOE)! 🌐 sites.google.com/view/findingth…

🖼️Reminder!! Finding the Frame will be held on August 5 at RL_Conference at University of Alberta! See you in Edmonton 📰Submit here: openreview.net/group?id=rl-co… ⌛️Deadline: May 30 🔗More info: sites.google.com/view/findingth…

So excited to get a chance to speak about this paper, the culmination of years of work from myself and many collaborators: arxiv.org/abs/2502.03349. Come join us and learn about the first RL agent that zero-shot aces driving benchmarks

🆕at Finding the Frame this year: outstanding paper awards!🏆 We will be highlighting exceptional papers across a few categories, including Most Thought-Provoking🤔 You still have until May 30 to submit- see our website for all the details! sites.google.com/view/findingth…

Anyone else finding the NeurIPS reviewer biding impossible this year? The affinity score is a poor proxy for my ability to review a paper, and having only the first 90 by that metric leaves papers mostly requiring significant background I am not particularly close to having.

I can highly recommend Stuart Russell's book "Human Compatible: Artificial intelligence and the Problem of Control", in particular "Chapter 6: The Not-So-Great AI Debate". Thanks Michael Dennis for recommending it!

2025 is the year of open-endedness. Delighted to be giving a talk at RAAIS in a couple of weeks’ time!

Excited to announce my first paper, with Jakob Foerster and Foerster Lab for AI Research, was accepted into @rl_conference 2025! We establish a new UED method called NCC that obtains strong performance based on principles of optimisation theory.

This is really exciting work! It's the first paper to take the convergence of UED algorithms seriously enough to provide a rigorous formal proof. Even better that it leads to a method with promising empirical results!

Unsupervised environment design (thus far had two kinds of papers: those that have beautiful theory and those that have convincing empirical results. Here's one that combines the two :)

🚨 We extended Finding the Frame's submission deadline to June 15 AoE! 🚨 ✨We're looking for bold ideas that rethink the foundations of RL: goals, values, rewards, formalisms, and beyond🚀 More details: sites.google.com/view/findingth… See you RL_Conference !

Now you have more time ! Got ideas that challenge RL norms ? This workshop is for you. You'd also be submitting to reviewers who (according to a previous iteration) find deep joy in reviewing for this workshop.

June 15th is Father's Day in the United States. We're co-opting the holiday to perhaps give you all a bit more time to prepare your papers for our RL_Conference workshop.

Some personal news: I joined Google DeepMind in Tim Rocktäschel's uber talented Open-Endedness team. I couldn’t be more excited for what we’re cooking. AI is the least open-ended it will ever be. Meta, it’s been a blast, an honor, and a privilege. I’m very grateful for the freedom and

Someone needs to use this as the basis of an unsupervised environment design algorithm to give AI designers direct control over agent behavior

Had a great time at RLDM this past week! Enjoyed many great conversations and fascinating talks Ben Eysenbach, Michael Dennis, Dan Haramati, DeepFlow. Thanks to The Alan Turing Institute and UKRI CDT in Foundational AI for the support!

Join the first seminar in our new 'Updates in Cooperative AI' series exploring our recent 'Multi-Agent Risks' report, live June 26th (5pm BST, 9am PT, 12pm ET) alongside Lewis Hammond Gillian Hadfield Michael Dennis.