38 citations · 51 across the 26 of their papers we have counts for
9 papers · 1 filter
Understanding the Staged Dynamics of Transformers in Learning Latent Structure
Rohan Saha, Farzane Aminmansour, Alona Fyshe
Language modeling has shown us that transformers can discover latent structure from context, but the dynamics of how they acquire different components of that structure remain poor…
Offline Preference Optimization via Maximum Marginal Likelihood Estimation
Saeed Najafi, Alona Fyshe
Aligning Large Language Models (LLMs) with human preferences is crucial, but standard methods like Reinforcement Learning from Human Feedback (RLHF) are often complex and unstable.…
Exploring Curriculum Learning for Vision-Language Tasks: A Study on Small-Scale Multimodal Training
Rohan Saha, Abrar Fahim, Alona Fyshe +1
For specialized domains, there is often not a wealth of data with which to train large machine learning models. In such limited data / compute settings, various methods exist aimin…
Finding Shared Decodable Concepts and their Negations in the Brain
Cory Efird, Alex Murphy, Joel Zylberberg +1
Prior work has offered evidence for functional localization in the brain; different anatomical regions preferentially activate for certain types of visual input. For example, the f…
Different Spectral Representations in Optimized Artificial Neural Networks and Brains
Richard C. Gerum, Cassidy Pirlot, Alona Fyshe +1
Recent studies suggest that artificial neural networks (ANNs) that match the spectral properties of the mammalian visual cortex -- namely, the eigenspectrum of the covar…
Long Term Fairness for Minority Groups via Performative Distributionally Robust Optimization
Liam Peet-Pare, Nidhi Hegde, Alona Fyshe
Fairness researchers in machine learning (ML) have coalesced around several fairness criteria which provide formal definitions of what it means for an ML model to be fair. However,…
No More Pesky Hyperparameters: Offline Hyperparameter Tuning for RL
Han Wang, Archit Sakhadeo, Adam White +7
The performance of reinforcement learning (RL) agents is sensitive to the choice of hyperparameters. In real-world settings like robotics or industrial control systems, however, te…
Resonance in Weight Space: Covariate Shift Can Drive Divergence of SGD with Momentum
Kirby Banman, Liam Peet-Pare, Nidhi Hegde +2
Most convergence guarantees for stochastic gradient descent with momentum (SGDm) rely on iid sampling. Yet, SGDm is often used outside this regime, in settings with temporally corr…
Maxmin Q-learning: Controlling the Estimation Bias of Q-learning
Qingfeng Lan, Yangchen Pan, Alona Fyshe +1
Q-learning suffers from overestimation bias, because it approximates the maximum action value using the maximum estimated action value. Algorithms have been proposed to reduce over…