A Reduction-based Framework for Sequential Decision Making with Delayed Feedback

TL;DR

This paper introduces a reduction-based framework that transforms algorithms for immediate feedback into sample-efficient solutions capable of handling stochastic delays in multi-agent sequential decision making, including bandits, MDPs, and MGs.

Contribution

The authors propose a novel reduction-based framework that extends existing algorithms to effectively manage stochastic delays in various sequential decision-making settings.

Findings

Framework matches or improves existing results for bandits, MDPs, and MGs.

First studies on delays in decision making with function approximation.

Provides sharp, comprehensive results for multi-agent delayed feedback scenarios.

Abstract

We study stochastic delayed feedback in general multi-agent sequential decision making, which includes bandits, single-agent Markov decision processes (MDPs), and Markov games (MGs). We propose a novel reduction-based framework, which turns any multi-batched algorithm for sequential decision making with instantaneous feedback into a sample-efficient algorithm that can handle stochastic delays in sequential decision making. By plugging different multi-batched algorithms into our framework, we provide several examples demonstrating that our framework not only matches or improves existing results for bandits, tabular MDPs, and tabular MGs, but also provides the first line of studies on delays in sequential decision making with function approximation. In summary, we provide a complete set of sharp results for multi-agent sequential decision making with delayed feedback.