Three Dogmas of Reinforcement Learning

TL;DR

This paper critically examines three foundational beliefs in reinforcement learning, questioning their roles and advocating for a paradigm shift towards more nuanced and agent-focused approaches.

Contribution

It challenges three core dogmas of reinforcement learning and proposes a rethinking of the field's foundational assumptions for future progress.

Findings

Questioning the environment-centric view shifts focus to agents.

Reconsidering the solution-centric view emphasizes adaptation over static solutions.

A nuanced approach to the reward hypothesis broadens goal representation.

Abstract

Modern reinforcement learning has been conditioned by at least three dogmas. The first is the environment spotlight, which refers to our tendency to focus on modeling environments rather than agents. The second is our treatment of learning as finding the solution to a task, rather than adaptation. The third is the reward hypothesis, which states that all goals and purposes can be well thought of as maximization of a reward signal. These three dogmas shape much of what we think of as the science of reinforcement learning. While each of the dogmas have played an important role in developing the field, it is time we bring them to the surface and reflect on whether they belong as basic ingredients of our scientific paradigm. In order to realize the potential of reinforcement learning as a canonical frame for researching intelligent agents, we suggest that it is time we shed dogmas one and two entirely, and embrace a nuanced approach to the third.