Learning optimal actions with imperfect images

Song Jun Park; Dale R. Shires

doi:10.1117/12.2518921

4 June 2019 Learning optimal actions with imperfect images

Song Jun Park, Dale R. Shires

Proceedings Volume 10996, Real-Time Image Processing and Deep Learning 2019; 109960F (2019) https://doi.org/10.1117/12.2518921
Event: SPIE Defense + Commercial Sensing, 2019, Baltimore, MD, United States

Abstract

Deep reinforcement learning has been successful in training an agent to play at human-level in Atari games. Here, outputs of the game images were fed into a deep neural network to compute optimal actions. Conceptually, reinforcement learning can be viewed as the intersection of planning, uncertainty, and learning. In this paper, deep reinforcement learning method is applied to solve a problem formulated as a partially observable Markov decision process (POMDP). Specifically, the input images are perturbed to introduce imperfect knowledge. POMDP formulations assume uncertainties in the true state space and thus a more accurate representation of the real-world scenarios. The deep Q-network is adopted to see if an optimal sequence of actions can be learned when the inputs are not fully observable. Experimental results indicated that optimal strategies were discovered by deep reinforcement learning in majority of test cases, albeit slower to converge to the optimal solution.

Conference Presentation

Citation Download Citation

Song Jun Park and Dale R. Shires "Learning optimal actions with imperfect images", Proc. SPIE 10996, Real-Time Image Processing and Deep Learning 2019, 109960F (4 June 2019); https://doi.org/10.1117/12.2518921

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available