Darshan Patil

What is the best way of switching between controllers in Reset Free RL? Read this thread (and come by our poster this Friday morning at ICLR, Hall B) to find out! #ICLR2024 1/n

In my lab, we have not one but four open postdoc positions! These positions cover developing foundation models for text, proteins, small molecules, genomic data, time series data, and astrophysics data! If you have strong research expertise and a PhD in LLMs and Foundation…

🥳 New Paper @ ACL Findings 🇦🇹 Instead of reverse engineering mechanisms in LLMs, can we inject our own known mechanism into a pretrained language model? Yes we can!

Transformers pre-trained on raw bytes (no tokenization) are SOTA lossless compressors (better than gzip, etc) on multiple data modalities (audio, images, text) With @HeurtelDepeiges @JoelVeness65957 Tim Genewein 📅 Tue 15 July ⏰ 16:30 – 19:00 📍East Exhibition Hall A-B #E-3410

Applying RL to a real system? Having insightful experience, new methods, reviews, analysis to share?We are excited to announce the @RL_Conference 2025 workshop on Practical Insights into RL for Real Systems! rl4rs.github.io/RL4RS

2025 BERT is NeoBERT! We have fully pre-trained a next-generation encoder for 2.1T tokens with the latest advances in data, training, and architecture. This is a heroic effort from my PhD student @lo_LB_La in collaboration with @qfournier2 and Mariam El Mezouar (1/n)

🚀Excited to share SFM– a method for IRL by direct policy optimization through a successor feature matching loss. Incredible collaboration with @harwiltz, @JesseFarebro, @irinarish, @GlenBerseth, and @sanjibac. Paper: arxiv.org/abs/2411.07007 Code: github.com/arnavkj1995/SFM 🧵⬇️

At @ChandarLab, we are happy to announce the second edition of our assistance program to provide feedback for members of communities underrepresented in AI who want to apply to high-profile graduate programs. Want feedback? Details: chandar-lab.github.io/grad-app-help/. Deadline: Nov 15!

🚨 New paper alert! [NeurIPS 2024 spotlight] 🚨 Trajectory Flow Matching with Applications to Clinical Time Series Modeling ⏳📈 With: @yuanpu__ , @YukiKawamura_ , Andrew Loza, @Yoshua_Bengio, @dlshung, @AlexanderTong7 💻: github.com/nZhangx/Trajec… 📄: arxiv.org/abs/2410.21154 🧵👇

Want to join @ChandarLab and @Mila_Quebec? I will be recruiting several graduate students (both MSc and PhD) for Fall 2025! Please apply through the Mila supervision matching process! The deadline to apply is 01 December!

Mila's annual supervision request process opens on October 15 to receive MSc and PhD applications for Fall 2025 admission! Join our community! More information here mila.quebec/en/prospective…

Our oral presentation at @iclr_conf is happening at 3:45 in Hall A2! Discover how we achieved superhuman performance on memory tasks. Missed it? Catch us at a poster session in Hall B #183 from 4:30 to 6:30.

Did you know that training an agent ensemble in RL (e.g. Bootstrapped DQN) can even hurt performance? In our ICLR 2024 paper “The Curse of Diversity in Ensemble-Based Exploration”, we argue that this counter-intuitive phenomenon is actually not that surprising once you realize…

Happening now! Come on by for a chat!

Making RL algorithms that work in the real world is tricky. In this new work we show how curriculums improve learning speed by focusing on collecting experience at the boundaries of the agent's capabilities. Find @dapatil211 at @iclr_conf in Friday am Halle B #155 4 more details.

I may have broken my foot, but our research poster is still kickin! Come see a unique one-legged presentation today at 4:30pm poster #77