Same State, Different Task: Continual Reinforcement Learning without Interference

TL;DR

This paper introduces OWL, a method for continual reinforcement learning that prevents interference between tasks by using separate policy heads and bandit-based policy selection, outperforming existing replay methods.

Contribution

The paper formalizes interference as distinct from forgetting and proposes OWL, a factorized policy approach with bandit-based selection to address interference in continual RL.

Findings

OWL outperforms existing replay-based CL methods in multiple RL environments.

OWL effectively prevents interference between incompatible tasks.

Bandit-based policy selection enables optimal task-specific policy reuse.

Abstract

Continual Learning (CL) considers the problem of training an agent sequentially on a set of tasks while seeking to retain performance on all previous tasks. A key challenge in CL is catastrophic forgetting, which arises when performance on a previously mastered task is reduced when learning a new task. While a variety of methods exist to combat forgetting, in some cases tasks are fundamentally incompatible with each other and thus cannot be learnt by a single policy. This can occur, in reinforcement learning (RL) when an agent may be rewarded for achieving different goals from the same observation. In this paper we formalize this "interference" as distinct from the problem of forgetting. We show that existing CL methods based on single neural network predictors with shared replay buffers fail in the presence of interference. Instead, we propose a simple method, OWL, to address this challenge. OWL learns a factorized policy, using shared feature extraction layers, but separate heads, each specializing on a new task. The separate heads in OWL are used to prevent interference. At test time, we formulate policy selection as a multi-armed bandit problem, and show it is possible to select the best policy for an unknown task using feedback from the environment. The use of bandit algorithms allows the OWL agent to constructively re-use different continually learnt policies at different times during an episode. We show in multiple RL environments that existing replay based CL methods fail, while OWL is able to achieve close to optimal performance when training sequentially.