An Initial Attempt of Combining Visual Selective Attention with Deep Reinforcement Learning

by   Liu Yuezhang, et al.
The University of Texas at Austin
Tsinghua University

Visual attention serves as a means of feature selection mechanism in the perceptual system. Motivated by Broadbent's leaky filter model of selective attention, we evaluate how such mechanism could be implemented and affect the learning process of deep reinforcement learning. We visualize and analyze the feature maps of DQN on a toy problem Catch, and propose an approach to combine visual selective attention with deep reinforcement learning. We experiment with optical flow-based attention and A2C on Atari games. Experiment results show that visual selective attention could lead to improvements in terms of sample efficiency on tested games. An intriguing relation between attention and batch normalization is also discovered.


page 2

page 4


Selective Particle Attention: Visual Feature-Based Attention in Deep Reinforcement Learning

The human brain uses selective attention to filter perceptual input so t...

Deep Reinforcement Learning Boosted by External Knowledge

Recent improvements in deep reinforcement learning have allowed to solve...

A Review of Uncertainty for Deep Reinforcement Learning

Uncertainty is ubiquitous in games, both in the agents playing games and...

Improving Sample Efficiency of Value Based Models Using Attention and Vision Transformers

Much of recent Deep Reinforcement Learning success is owed to the neural...

Rainbow: Combining Improvements in Deep Reinforcement Learning

The deep reinforcement learning community has made several independent i...

Learning to attend in a brain-inspired deep neural network

Recent machine learning models have shown that including attention as a ...

Please sign up or login with your details

Forgot password? Click here to reset