#deeperlearningblog 検索結果

Kempner Institute at Harvard University

/07/14

New in the #DeeperLearningBlog: Researchers from the #KempnerInstitute, @Cornell and @CMU_Robotics introduce a new method for improving offline RL by scaling-up test-time compute. kempnerinstitute.harvard.edu/research/deepe… #AI #RL (1/2)

KempnerInst's tweet card. This research introduces a novel approach to scaling reinforcement learning (RL) during training and inference. Inspired by the recent work on LLM test-time scaling, we demonstrate how greater...

Scaling Offline Reinforcement Learning at Test Time - Kempner Institute

ソース: kempnerinstitute.harvard.edu

Kempner Institute at Harvard University

@KempnerInst

/07/28

New in the #DeeperLearningBlog: the #KempnerInstitute's @maryiletey presents work recently published in PNAS that offers generalizable insights into in-context learning (ICL) in an analytically-solvable model architecture. bit.ly/4lPK15p #AI @PNASNews (1/2)

KempnerInst's tweet card. Attention-based architectures are a powerful force in modern AI. In particular, the emergence of in-context learning enables these models to perform tasks far beyond the original next-token predict...

Solvable Model of In-Context Learning Using Linear Attention - Kempner Institute

ソース: kempnerinstitute.harvard.edu

Kempner Institute at Harvard University

@KempnerInst

/07/22

New in the #DeeperLearningBlog: #KempnerInstitute research fellow @t_andy_keller introduces the first flow equivariant neural networks, which reflect motion symmetries, greatly enhancing generalization and sequence modeling. bit.ly/451fQ48 #AI #NeuroAI

KempnerInst's tweet card. Sequence transformations, like visual motion, dominate the world around us, but are poorly handled by current models. We introduce the first flow equivariant models that respect these motion symmet...

Flow Equivariant Recurrent Neural Networks - Kempner Institute

ソース: kempnerinstitute.harvard.edu

Kempner Institute at Harvard University

@KempnerInst

/07/18

New in the #DeeperLearningBlog: #KempnerInstitute researchers @WangBinxu and @johnjvastola explain their work uncovering the linear Gaussian structure in diffusion models and the potential to use it to enhance performance. bit.ly/4lCauDv #AI #DiffusionModels

Shun Li

@shunnnli

/08/04

Check out the #DeeperLearningBlog that summarizes key findings of our recent preprint on sign switching synapses! (with my speculations on its connections with AI 🤖)

Kempner Institute at Harvard University

@KempnerInst

/08/04

New in the #DeeperLearningBlog: @shunnnli describes the discovery of a type of synapse that switches between more excitatory and more inhibitory in an experience-dependent manner. bit.ly/40OYYw4 #neuroscience #dopamine #RL @blsabatini

KempnerInst's tweet card. Synaptic plasticity rules in the brain are normally thought of as changing synaptic weights but not signs, unlike artificial neural networks. We show that a type of synapse in the […]

ANN-like Synapses in the Brain Mediate Online Reinforcement Learning - Kempner Institute

ソース: kempnerinstitute.harvard.edu

Kempner Institute at Harvard University

@KempnerInst

/06/27

New in the #DeeperLearningBlog: Kempner researchers Nikhil Anand (@nikhil_anand91) and Chloe Su (@Huangyu58589918) discuss new work on how numerical precision can impact the accuracy and stability of #LLMs. kempnerinstitute.harvard.edu/research/deepe… #AI (1/2)

KempnerInst's tweet card. This research uncovers consistent training instabilities when using new, highly efficient low-precision formats, which has implications for the development of next-generation AI. By pinpointing the...

Characterization and Mitigation of Training Instabilities in Microscaling Formats - Kempner...

ソース: kempnerinstitute.harvard.edu

Kempner Institute at Harvard University

@KempnerInst

/07/30

New in the #DeeperLearningBlog: @GaoZhaolin and collaborators including the #KempnerInstitute's Kianté Brantley presents a powerful new #RL algorithm tailored for reasoning tasks with #LLMs that updates using only one generation per prompt. bit.ly/44US1Mt @xkianteb #AI

KempnerInst's tweet card. Recent LLM advances show the effectiveness of RL with rule-based rewards, but methods like GRPO and PPO are costly due to critics or multiple generations per prompt. We propose a […]

Accelerating RL for LLM Reasoning with Optimal Advantage Regression - Kempner Institute

ソース: kempnerinstitute.harvard.edu

Kempner Institute at Harvard University

@KempnerInst

/08/04

ANN-like Synapses in the Brain Mediate Online Reinforcement Learning - Kempner Institute

ソース: kempnerinstitute.harvard.edu

Shun Li

@shunnnli

/08/04

Check out the #DeeperLearningBlog that summarizes key findings of our recent preprint on sign switching synapses! (with my speculations on its connections with AI 🤖)