Dylan R. Ashley

Using only box-forwarding speed as the reward, our Stackelberg PPO automatically evolves robots with arms for pushing and legs for moving. The key idea is a novel game-theoretic view of structure–control co-design, yielding more effective optimization and dramatically better designs. Come see our poster at ICLR 2026 on Apr 25, 10:30 AM, at P4-#4810. With @YuhuiWangAI, @YanningD_AI, @oneDylanAshley. Paper: arxiv.org/abs/2603.15388 Project Page: yanningdai.github.io/stackelberg-pp…

Do you like RL and math? Our collaboration, IDSIA-KAUST-NNAISENSE, has the most detailed exploration of the convergence and stability of modern RL frameworks like Upside-Down RL, Online Decision Transformers, and Goal-Conditioned Supervised Learning arxiv.org/abs/2502.05672

Are you a rising star in AI? 🌟 Join us as a speaker for the 4th edition of the KAUST Rising Stars in AI Symposium. In the past 2 years co-organizing this event, I've met incredible researchers now in top industrial and academic positions worldwide. More info: 📅 Event date: April 7-10, 2025 ⏳ Application deadline: December 18 🔗 Apply here: kaust.edu.sa/en/news/rising…

Excited to share new work done with @oneDylanAshley, Zachary Friggstad and @SchmidhuberAI on stories, music, movies and information theory, published in IEEE TPAMI📚🎶 🎬🤖 Below a quick summary and a web tool you can use to sort your playlists:) 1/n doi.org/10.1109/TPAMI.…

Some very cool #AI #DeepLearning #RL work to have been involved with. Interestingly, there's no clear bound on the depth of the network or general scalability here, so a lot of potential.

🔔 new 𝗔𝗴𝗲𝗻𝘁-𝗮𝘀-𝗮-𝗝𝘂𝗱𝗴𝗲 paper: 𝗖𝗮𝗻 𝗔𝗜 𝗮𝗴𝗲𝗻𝘁𝘀 𝗲𝘃𝗮𝗹𝘂𝗮𝘁𝗲 𝗔𝗜 𝗮𝗴𝗲𝗻𝘁𝘀 𝗮𝘀 𝗲𝗳𝗳𝗲𝗰𝘁𝗶𝘃𝗲𝗹𝘆 𝗮𝘀 𝗵𝘂𝗺𝗮𝗻𝘀? 𝗬𝗲𝘀, 𝘁𝗵𝗲𝘆 𝗰𝗮𝗻! 📄 arxiv.org/abs/2410.10934… 👨‍💻 github.com/metauto-ai/age… Introducing 𝗔𝗴𝗲𝗻𝘁-𝗮𝘀-𝗮-𝗝𝘂𝗱𝗴𝗲, a groundbreaking proof-of-concept that reduces costs and time by 97%, while providing rich, intermediate feedback. It precisely captures the natural step-by-step processes of agentic systems. We also developed 𝗗𝗲𝘃𝗔𝗜, a new benchmark featuring 55 automated AI development tasks and 365 requirements. Agent-as-a-Judge not only outperforms LLM-as-a-Judge but also closely mirrors human evaluations with greater efficiency and precision. The real game-changer? It provides reliable reward signals, paving the way for scalable, self-improving agentic systems. Thanks my Meta/KAUST mentors/peers/collaborators @SchmidhuberAI @tydsh @zechunliu @vikasc @YoungXiong1 @vikasc @Obs01ete @erniecyc @oneDylanAshley ...

Can neural networks with 5000 layers improve long-term planning? 🤖 Check out our latest research with @SchmidhuberAI, @oneDylanAshley, and team: arxiv.org/abs/2406.08404 #AI #DeepLearning #RL