Charles P. Rizzo, Catherine D. Schuman, and James S. Plank

April, 2023

Neuro-Inspired Computational Elements Conference

https://dl.acm.org/doi/proceedings/10.1145/3584954

View Article

Abstract

In this work, we address the problem of training a neuromorphic agent to work on data from event-based cameras. Although event-based camera data is much sparser than standard video frames, the sheer number of events can make the observation space too complex to effectively train an agent. We construct multiple neuromorphic networks that downsample the camera data so as to make training more effective. We then perform a case study of training an agent to play the Atari Pong game by converting each frame to events and downsampling them. The final network combines both the downsampling and the agent. We discuss some practical considerations as well.

Citation Information

Text

author      C. P. Rizzo and C. D. Schuman and J. S. Plank
title       Neuromorphic Downsampling of Event-Based Camera Output
booktitle   Proceedings of the 2023 Annual Neuro-Inspired Computational Elements Conference
publisher   ACM
doi         10.1145/3584954.3584962
url         https://dl.acm.org/doi/abs/10.1145/3584954.3584962
month       April
year        2023

Bibtex

@INPROCEEDINGS{rsp:23:nde,
    author = "C. P. Rizzo and C. D. Schuman and J. S. Plank",
    title = "Neuromorphic Downsampling of Event-Based Camera Output",
    booktitle = "Proceedings of the 2023 Annual Neuro-Inspired Computational Elements Conference",
    publisher = "ACM",
    doi = "10.1145/3584954.3584962",
    url = "https://dl.acm.org/doi/abs/10.1145/3584954.3584962",
    month = "April",
    year = "2023"
}