David Abel

New #RLC2024 paper Three Dogmas of Reinforcement Learning joint w/ @mark_ho_ and @aharutyu! arxiv.org/pdf/2407.10583 We reflect on where our scientific paradigm needs adjustment, and suggest three departures from previous conventions. Curious to hear what folks think! 🧵

Excited about this new work led by @jonathanrichens, joint with @alexis_bellot_ and @tom4everitt Main result: Any agent that can solve a sufficiently rich set of goal-directed tasks must have learned a predictive model of the environment arxiv.org/abs/2506.01622

Are world models necessary to achieve human-level agents, or is there a model-free short-cut? Our new #ICML2025 paper tackles this question from first principles, and finds a surprising answer, agents _are_ world models… 🧵

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

🚨 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…

🚨 Excited to share my new paper with @Dr_Atoosa “Characterizing AI Agents for Alignment and Governance” We identify four key dimensions of AI agency—autonomy, efficacy, goal complexity & generality—and use this analysis to construct “agentic profiles” for a range of AI agents.

paper: arxiv.org/abs/2503.06343 project page: github.com/francelico/dea…

Looking forward to presenting our study on representation learning for on-policy actor-critic algorithms in Singapore for #ICLR2025 ! Here's a picture to tease 3 key insights in our study... 1⃣ Decoupled actor-critic model architectures outperform their shared counterparts.…

After last year's success, Finding the Frame is back at #RLC2025 🚀! Do you want to rethink the conceptual foundations of RL? Define new problems, challenge assumptions, or test the limits of RL paradigms? Join us! Submit by: 30th May 2025, AoE Guidelines: sites.google.com/view/findingth…

Ever wonder how extremely rare, high-risk black swan events can sneak up even when nothing changes? In our #ICLR2025 paper, we introduce S-BLACK SWAN: a framework showing how misperceived rewards and probabilities alone can spawn catastrophic surprises—no shifting environment…

✨Recordings of Finding the Frame Workshop at @RL_Conference 2024 are now up on our website!✨ sites.google.com/view/findingth…

Save the date! RLDM 2025, The Multi-disciplinary Conference on Reinforcement Learning and Decision Making, is only around the corner. Visit our website to keep an eye on our submission deadlines👀 rldm.org

📢 Exciting News! The Fourth Conference on Lifelong Learning Agents (CoLLAs 2025) will be held at the University of Pennsylvania (@Penn) in Philadelphia, USA 🇺🇸 🗓️ Important Dates: Abstract Deadline: Feb 21, 2025 Submission Deadline: Feb 26, 2025 Conference Dates: Aug 11 - Aug…

The team @jhamrick and I co-lead is hiring a research engineer. If you are interested in improving the capabilities of LLMs in the planning and reasoning space, and building generally capable agents, please apply! boards.greenhouse.io/deepmind/jobs/…

New conference on safe and ethical AI! Our goal is to convene a broad group of experts from academia, civil society, industry, media, and governments to discuss the latest developments in AI safety and ethics. Please apply here by Nov 24: iaseai.org/conference/app…

Reinforcement learning in #AI is hard, so I’ve made a website to collect answers I’ve given to common RL questions. It's named Decisions & Dragons. It’s launching with 8 questions and answers, but I will add to it in the future. A 🧵to give a preview with the link below.

🚀 I am recruiting PhD students for Fall 2025 at the UCLA Robot Intelligence Lab! 🤖 If you are interested in robot learning and human-robot interaction, mark me as a potential adivisor when you apply to the UCLA CS PhD program! #PhD #Robotics @CS_UCLA

🎇 I’m on the academic job market! I’m a PhD candidate at @mldcmu. My research tackles challenges that arise from the sequential nature of human-AI interaction. Toward this goal, my work involves: 🤖 reinforcement learning, 🧠 foundation models, and 👩‍💻 human-centered AI.…

I've just published a paper positing a new kind of fundamental physical law bounding the rate at which any physical quantity can grow or converge. 1/🧵

BREAKING NEWS The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Chemistry with one half to David Baker “for computational protein design” and the other half jointly to Demis Hassabis and John M. Jumper “for protein structure prediction.”

With one of our favorite talk titles, Finale Doshi-Velez told us about the ways her RL work is impacting medicine + how assuming human involvement leads to new algorithms (Link below)