We view large language models (LLMs) as stochastic language layers in
a ...
The growing utilization of machine learning (ML) in decision-making proc...
Actor-critic (AC) methods are widely used in reinforcement learning (RL)...
We consider minimizing functions for which it is expensive to compute th...
Parameter-efficient fine-tuning (PEFT) methods can adapt large language
...
We use functional mirror ascent to propose a general framework (referred...
We investigate the impact of aliasing on generalization in Deep Convolut...
We study the stochastic bilinear minimax optimization problem, presentin...
This work revisits the use of information criteria to characterize the
g...
Most stochastic optimization methods use gradients once before discardin...
Tail averaging consists in averaging the last examples in a stream. Comm...
Despite many algorithmic advances, our theoretical understanding of prac...
The loss function of deep networks is known to be non-convex but the pre...
This paper proposes a new approach to representation learning based on
g...
We establish geometric and topological properties of the space of value
...
Entropy regularization is commonly used to improve policy optimization i...
Entropy regularization is commonly used to improve policy optimization i...
Our goal is to improve variance reducing stochastic methods through bett...
We provide a comparative study of several widely used off-policy estimat...
We tackle the issue of finding a good policy when the number of policy
u...
The standard approach to supervised classification involves the minimiza...
We propose the stochastic average gradient (SAG) method for optimizing t...
We propose an extension of the Restricted Boltzmann Machine (RBM) that a...