Proposal for Cognitive Architecture and Transformer Integration: Online Learning from Agent Experience

Authors

  • John E. Laird IQMRI: Center for Integrated Cognition
  • Robert E. Wray IQMRI: Center for Integrated Cognition
  • Steven Jones IQMRI: Center for Integrated Cognition
  • James R. Kirk IQMRI: Center for Integrated Cognition
  • Peter Lindes IQMRI: Center for Integrated Cognition

DOI:

https://doi.org/10.1609/aaaiss.v2i1.27692

Keywords:

Cognitive Architecture, Generative Model, Transformer, Soar

Abstract

We explore the potential integration of Transformers, trained online in real-time using an agent's ongoing experiences, as a learning and memory component of a cognitive architecture such as Soar. We identify key challenges and potential capabilities enabled by such an integration.
AAAI Fall Symposium 2023 Proceedings Cover

Downloads

Published

2024-01-22

How to Cite

Laird, J. E., Wray, R. E., Jones, S., Kirk, J. R., & Lindes, P. (2024). Proposal for Cognitive Architecture and Transformer Integration: Online Learning from Agent Experience. Proceedings of the AAAI Symposium Series, 2(1), 302–306. https://doi.org/10.1609/aaaiss.v2i1.27692

Issue

Vol. 2 No. 1: Proceedings of the 2023 AAAI Fall Symposia

Section

Integration of Cognitive Architectures and Generative Models