Most language models (LMs) are trained and applied in an autoregressive
...
Contextual bandit algorithms are ubiquitous tools for active sequential
...
Virtual support agents have grown in popularity as a way for businesses ...
We develop confidence bounds that hold uniformly over time for off-polic...
We apply empirical likelihood techniques to contextual bandit policy val...
In this work, we describe practical lessons we have learned from success...
We create a new online reduction of multiclass classification to binary
...
We propose a general framework for sequential and dynamic acquisition of...
Extreme classification problems are multiclass and multilabel classifica...
We present RandomizedCCA, a randomized algorithm for computing canonical...
This work provides simple algorithms for multi-class (and multi-label)
p...
Representing examples in a way that is compatible with the underlying
cl...
We propose a sampling scheme suitable for reducing a data set prior to
s...
We address the problem of learning in an online setting where the learne...