A Minimal “Functionally Sentient” Organism Trained With Backpropagation Through Time
This article presents a scenario where a simple simulated organism must explore and exploit an environment containing a food pile. The organism learns to make observations of the environment, use memory to record those observations, and thus plan and navigate to the regions with the strongest food d...
Saved in:
Published in: | Adaptive behavior 2023-12, Vol.31 (6), p.531-544 |
---|---|
Main Authors: | , , |
Format: | Article |
Language: | eng |
Online Access: | Request full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | This article presents a scenario where a simple simulated organism must explore and exploit an environment containing a food pile. The organism learns to make observations of the environment, use memory to record those observations, and thus plan and navigate to the regions with the strongest food density. We compare different reinforcement learning algorithms with an adaptive dynamic programming algorithm and conclude that backpropagation through time can convincingly solve this recurrent neural-network challenge. Furthermore, we argue that this algorithm successfully mimics a minimal ‘functionally sentient’ organism’s fundamental objectives and mental environmental-mapping skills while seeking a food pile distributed statically or randomly in an environment. |
---|---|
ISSN: | 1059-7123 1741-2633 |