4 papers
Inhibitory normalization of error signals improves learning in neural circuits
Roy Henha Eyono, Daniel Levenstein, Arna Ghosh +2
Normalization is a critical operation in neural circuits. In the brain, there is evidence that normalization is implemented via inhibitory interneurons and allows neural population…
Tracing the Representation Geometry of Language Models from Pretraining to Post-training
Melody Zixuan Li, Kumar Krishna Agrawal, Arna Ghosh +4
Standard training metrics like loss fail to explain the emergence of complex capabilities in large language models. We take a spectral approach to investigate the geometry of learn…
Why all roads don't lead to Rome: Representation geometry varies across the human visual cortical hierarchy
Arna Ghosh, Zahraa Chorghay, Shahab Bakhtiari +1
Biological and artificial intelligence systems navigate the fundamental efficiency-robustness tradeoff for optimal encoding, i.e., they must efficiently encode numerous attributes…
Learning Successor Features the Simple Way
Raymond Chua, Arna Ghosh, Christos Kaplanis +2
In Deep Reinforcement Learning (RL), it is a challenge to learn representations that do not exhibit catastrophic forgetting or interference in non-stationary environments. Successo…