🔥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

Q. Can we solve learning dynamics of modern deep learning models trained on large datasets? A. Yes, by combining symmetry and modified equation analysis! co-led with Daniel Kunin (now on twitter) & Javier Sagastuy Surya Ganguli Daniel Yamins Neural Mechanics arxiv.org/abs/2012.04728 1/8

Very cool: multi-timescale learning in spiking networks improves performance and robustness in tasks with strong temporal structure.

Eight amazing speakers are presenting their ideas on Distributed Representations in RL and the brain on September 21. Very happy to organize this satellite workshop at Bernstein Network Computational Neuroscience with Paul Masset

Is this new 👀preprint👀 worth your time? ...that depends on what your time is worth! TL;DR We ask how time’s value is encoded in the brain and used in the neural dynamics of deliberation. We answer (& validate using NHP behaviour and recordings) with an adaptive strategy:

Enjoying the BCCN neuroAI workshop on RL today. Jan Drugowitsch showed ongoing work trying to discriminate different codes from neural data. This is a great topic we can't avoid in getting theory and experiment work together! x.com/mptouzel/statu…

Another #MyoChallenge top-submission for the `Baoding Ball` Task. This time from the stiff_fingers team (Pablo Tano Alberto Chiappa Nisheet Patel Alexandre Pouget Alexander Mathis) ! 1 week until the end of phase 1! 🏃

It has been an absolute pleasure collaborating with A. Mathis Group on this fun project! Very grateful to work with some awesome colleagues Alberto Chiappa, Nisheet Patel, Alexandre Pouget and Alexander Mathis

In San Diego for #SfN22 ! Looking forward to presenting "a distributional code for learning across timescales in dopamine-based reinforcement learning" on Sunday afternoon (poster board SS13) With Pablo Tano Athar N Malik, MD, PhD HyungGoo Kim @PolBech1 Alexandre Pouget Naoshige Uchida

Amazing work from Alberto Chiappa and the A. Mathis Group ! Robust locomotion is made possible by an attention mechanism that gates the proprioceptive state history of the agent's joints

Major breakthrough in recovery from paralysis! With the aid of brain-spine interface, this patient was able climb steps and navigate complex terrain. He still needs some support with crutches or a walker, but the recovery is astounding nature.com/articles/s4158…

Hey #SfN23 ! Want to hear about temporal representations, dopamine and reinforcement learning? Come see our poster TT26 on Wednesday 15th morning with Pablo Tano HyungGoo Kim Athar N Malik, MD, PhD Alexandre Pouget @naoshigeuchida #SfN2023

How can animals learn from rewards across timescales? Our work with Pablo Tano HyungGoo Kim Athar N Malik, MD, PhD Alexandre Pouget @naoshigeuchida explores how dopamine neurons could enable multi-timescale reinforcement learning in the brain biorxiv.org/content/10.110…

Are you interested in motor skills, musculoskeletal control and reinforcement learning? Check out our manuscript: biorxiv.org/content/10.110…

How does the brain control the numerous muscles of the body? Let's say you want to rotate two balls in your hand, how does your brain achieve that? Read our article in Neuron to learn more! cell.com/neuron/fulltex…

Happy to share the latest version of our work on compositional cognitive maps in hippocampus, with Jo Warren, @jrcwhittington, Tim Behrens. We propose hippocampus constructs maps from cortical building blocks in replay – now with empirical support! nature.com/articles/s4159… 1/9

Our work with Pablo Tano , HyungGoo Kim Athar N Malik, MD, PhD Alexandre Pouget and Nao Uchida exploring how dopamine neurons could enable multi-timescale reinforcement learning in the brain is out in nature nature.com/articles/s4158…