Unsupervised Video Object Segmentation for Deep Reinforcement Learning
This addresses the challenge of sample efficiency and interpretability in deep reinforcement learning for practitioners, though it is incremental as it builds on existing methods by adding object detection.
The paper tackles the problem of improving deep reinforcement learning by automatically detecting and segmenting moving objects in an unsupervised manner, using structure from motion, which reduces the amount of environment interaction needed for policy learning and makes the resulting policies more interpretable.
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.