Unsupervised Video Object Segmentation for Deep Reinforcement Learning

TL;DR

This paper introduces MOREL, an unsupervised method that detects and segments moving objects in video to improve deep reinforcement learning policies, making them more interpretable and efficient.

Contribution

The paper presents a novel unsupervised approach for object detection in reinforcement learning, leveraging motion cues to enhance policy learning and interpretability.

Findings

Reduced environment interactions needed for good policies.

Enhanced interpretability of learned policies.

Effective detection of relevant moving objects in Atari games.

Abstract

We present a new technique for deep reinforcement learning that automatically detects moving objects and uses the relevant information for action selection. The detection of moving objects is done in an unsupervised way by exploiting structure from motion. Instead of directly learning a policy from raw images, the agent first learns to detect and segment moving objects by exploiting flow information in video sequences. The learned representation is then used to focus the policy of the agent on the moving objects. Over time, the agent identifies which objects are critical for decision making and gradually builds a policy based on relevant moving objects. This approach, which we call Motion-Oriented REinforcement Learning (MOREL), is demonstrated on a suite of Atari games where the ability to detect moving objects reduces the amount of interaction needed with the environment to obtain a good policy. Furthermore, the resulting policy is more interpretable than policies that directly map images to actions or values with a black box neural network. We can gain insight into the policy by inspecting the segmentation and motion of each object detected by the agent. This allows practitioners to confirm whether a policy is making decisions based on sensible information.