Machine Versus Human Attention In Deep Reinforcement Learning Tasks

Abstract

Deep reinforcement learning (RL) algorithms are powerful tools for solving visuomotor decision tasks. However, the trained models are often difficult to interpret, because they are represented as end-to-end deep neural networks. In this paper, we shed light on the inner workings of such trained models by analyzing the pixels that they attend to during task execution, and comparing them with the pixels attended to by humans executing the same tasks. To this end, we investigate the following two questions that, to the best of our knowledge, have not been previously studied. 1) How similar are the visual representations learned by RL agents and humans when performing the same task? and, 2) How do similarities and differences in these learned representations explain RL agents' performance on these tasks? Specifically, we compare the saliency maps of RL agents against visual attention models of human experts when learning to play Atari games. Further, we analyze how hyperparameters of the d

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyguo2020machine

Related papers

• Visualizing And Understanding Atari Agents (2017)0.00
• What Deep Reinforcement Learning Tells Us About Human Motor Learning And Vice-versa (2022)0.00
• Learn To Interpret Atari Agents (2018)0.00
• Visual Explanation Using Attention Mechanism In Actor-critic-based Deep Reinforcement Learning (2021)10.21
• Exploratory Not Explanatory: Counterfactual Analysis Of Saliency Maps For Deep Reinforcement Learning (2019)0.00
• Can You See How I Learn? Human Observers' Inferences About Reinforcement Learning Agents' Learning Processes (2025)0.00
• Revealing The Learning Process In Reinforcement Learning Agents Through Attention-oriented Metrics (2024)0.00
• Comparing Reinforcement Learning And Human Learning Using The Game Of Hidden Rules (2023)2.26