Satpreet Singh

1/n Excited to share our new preprint where we study turbulent plume tracking using deep reinforcement learning (DRL) trained RNN *agents* and find many intriguing similarities with flying insects. w/ @FlorisBreugel @RajeshPNRao @bingbrunton; #tweeprint @flypapers #Drosophila

If you like scaling laws in AI you’ll love scaling laws in biology, like allometric scaling of energy production density as a power law with exponent -1/4 across 100’s of millions of years of evolution. Also a new and similar scaling law for sleep - connecting it to metabolism!

Why do video models handle motion so poorly? It might be lack of motion equivariance. Very excited to introduce: Flow Equivariant RNNs (FERNNs), the first sequence models to respect symmetries over time. Paper: arxiv.org/abs/2507.14793 Blog: kempnerinstitute.harvard.edu/research/deepe… 1/🧵

NeurIPS is pleased to officially endorse EurIPS, an independently-organized meeting taking place in Copenhagen this year, which will offer researchers an opportunity to additionally present their accepted NeurIPS work in Europe, concurrently with NeurIPS. Read more in our blog…

Announcing AI for Science at @NeurIPSConf 2025! Join us in discussing the reach and limits of AI for scientific discovery 🚀 📍 Workshop submission deadline: Aug 22 💡 Dataset proposal competition: more details coming soon! ✨ Amazing line up of speakers and panelists

New Book & Video Series!!! (late 2025) Optimization Bootcamp: Applications in Machine Learning, Control, and Inverse Problems Comment for a sneak peak to help proofread and I'll DM (proof reading, typos, HW problems, all get acknowledgment in book!)

At ICML for the next 2 days to present multiple works, if you're into interpretability, complexity, or just wanna know how cool @KempnerInst is, hit me up 👋

Excited to share a new project spanning cognitive science and AI where we develop a novel deep reinforcement learning model---Multitask Preplay---that explains how people generalize to new tasks that were previously accessible but unpursued.

Everyone knows action chunking is great for imitation learning. It turns out that we can extend its success to RL to better leverage prior data for improved exploration and online sample efficiency! colinqiyangli.github.io/qc/ The recipe to achieve this is incredibly simple. 🧵 1/N

(1/7) New preprint from the Rajan Lab! 🧠🤖 @RyanPaulBadman1 & @SimmonsEdler show–through cog sci, neuro & ethology-how an AI agent with fewer ‘neurons’ than an insect can forage, find safety & dodge predators in a virtual world. Here's what we did Paper: arxiv.org/pdf/2506.06981

🚨 Registration is live! 🚨 The New England Mechanistic Interpretability (NEMI) Workshop is happening August 22nd 2025 at Northeastern University! A chance for the mech interp community to nerd out on how models really work 🧠🤖 🌐 Info: nemiconf.github.io/summer25/ 📝 Register:…

Watch the latest update from the Neuralink team.

As AI agents face increasingly long and complex tasks, decomposing them into subtasks becomes increasingly appealing. But how do we discover such temporal structure? Hierarchical RL provides a natural formalism-yet many questions remain open. Here's our overview of the field🧵

Now in Nature! 🚀 Our method learns a controllable 3D model of any robot from vision, enabling single-camera closed-loop control at test time! This includes robots previously uncontrollable, soft, and bio-inspired, potentially lowering the barrier of entry to automation! Paper:…

This collab was one of the most beautiful papers I've ever worked on! The amount I learned from @danielwurgaft was insane and you should follow him to inherit some gems too :D

🚨New paper! We know models learn distinct in-context learning strategies, but *why*? Why generalize instead of memorize to lower loss? And why is generalization transient? Our work explains this & *predicts Transformer behavior throughout training* without its weights! 🧵 1/

Humans and animals can rapidly learn in new environments. What computations support this? We study the mechanisms of in-context reinforcement learning in transformers, and propose how episodic memory can support rapid learning. Work w/ @KanakaRajanPhD: arxiv.org/abs/2506.19686

Agents training in RL gym

(1/n) Since its publication in 2017, PPO has essentially become synonymous with RL. Today, we are excited to provide you with a better alternative - EPO.

guys?

If you're at #RLDM2025, check out our contributed talk at Session 3 (Fri 6/13, 12:10pm), presented by my brilliant co-first-author on this project @sonjajohnsonyu! Wasn't able to make it in person, but would love to hear your thoughts @KempnerInst @harvardmed