Saved in:
Bibliographic Details
Main Authors: Alamdari, Parand A., Cao, Yanshuai, Wilson, Kevin H.
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.19317
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909369818415104
author Alamdari, Parand A.
Cao, Yanshuai
Wilson, Kevin H.
author_facet Alamdari, Parand A.
Cao, Yanshuai
Wilson, Kevin H.
contents We present substantial evidence demonstrating the benefits of integrating Large Language Models (LLMs) with a Contextual Multi-Armed Bandit framework. Contextual bandits have been widely used in recommendation systems to generate personalized suggestions based on user-specific contexts. We show that LLMs, pre-trained on extensive corpora rich in human knowledge and preferences, can simulate human behaviours well enough to jump-start contextual multi-armed bandits to reduce online learning regret. We propose an initialization algorithm for contextual bandits by prompting LLMs to produce a pre-training dataset of approximate human preferences for the bandit. This significantly reduces online learning regret and data-gathering costs for training such models. Our approach is validated empirically through two sets of experiments with different bandit setups: one which utilizes LLMs to serve as an oracle and a real-world experiment utilizing data from a conjoint survey experiment.
format Preprint
id arxiv_https___arxiv_org_abs_2406_19317
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Jump Starting Bandits with LLM-Generated Prior Knowledge
Alamdari, Parand A.
Cao, Yanshuai
Wilson, Kevin H.
Machine Learning
Artificial Intelligence
Computation and Language
We present substantial evidence demonstrating the benefits of integrating Large Language Models (LLMs) with a Contextual Multi-Armed Bandit framework. Contextual bandits have been widely used in recommendation systems to generate personalized suggestions based on user-specific contexts. We show that LLMs, pre-trained on extensive corpora rich in human knowledge and preferences, can simulate human behaviours well enough to jump-start contextual multi-armed bandits to reduce online learning regret. We propose an initialization algorithm for contextual bandits by prompting LLMs to produce a pre-training dataset of approximate human preferences for the bandit. This significantly reduces online learning regret and data-gathering costs for training such models. Our approach is validated empirically through two sets of experiments with different bandit setups: one which utilizes LLMs to serve as an oracle and a real-world experiment utilizing data from a conjoint survey experiment.
title Jump Starting Bandits with LLM-Generated Prior Knowledge
topic Machine Learning
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2406.19317