Explain Your Move: Understanding Agent Actions Using Focused Feature Saliency

by   Piyush Gupta, et al.
University of California, Irvine
IIT Kharagpur

As deep reinforcement learning (RL) is applied to more tasks, there is a need to visualize and understand the behavior of learned agents. Saliency maps explain agent behavior by highlighting the features of the input state that are most relevant for the agent in taking an action. Existing perturbation-based approaches to compute saliency often highlight regions of the input that are not relevant to the action taken by the agent. Our approach generates more focused saliency maps by balancing two aspects (specificity and relevance) that capture different desiderata of saliency. The first captures the impact of perturbation on the relative expected reward of the action to be explained. The second downweights irrelevant features that alter the relative expected rewards of actions other than the action to be explained. We compare our approach with existing approaches on agents trained to play board games (Chess and Go) and Atari games (Breakout, Pong and Space Invaders). We show through illustrative examples (Chess, Atari, Go), human studies (Chess), and automated evaluation methods (Chess) that our approach generates saliency maps that are more interpretable for humans than existing approaches.


page 3

page 5

page 6

page 8

page 13

page 14


Benchmarking Perturbation-based Saliency Maps for Explaining Deep Reinforcement Learning Agents

Recent years saw a plethora of work on explaining complex intelligent ag...

Exploratory Not Explanatory: Counterfactual Analysis of Saliency Maps for Deep RL

Saliency maps have been used to support explanations of deep reinforceme...

Free-Lunch Saliency via Attention in Atari Agents

We propose a new approach to visualize saliency maps for deep neural net...

Human versus Machine Attention in Deep Reinforcement Learning Tasks

Deep reinforcement learning (RL) algorithms are powerful tools for solvi...

Confounder-Aware Visualization of ConvNets

With recent advances in deep learning, neuroimaging studies increasingly...

Learning to predict where to look in interactive environments using deep recurrent q-learning

Bottom-Up (BU) saliency models do not perform well in complex interactiv...

Quality-Aware Multimodal Saliency Detection via Deep Reinforcement Learning

Incorporating various modes of information into the machine learning pro...

Code Repositories

Please sign up or login with your details

Forgot password? Click here to reset