Oak Lab

The goal of Oak Lab is to discover and implement the fundamental computational principles that allow agents to learn to achieve their goals in complex habitats. We take the big world hypothesis (Javed and Sutton, 2024) as a given, and we want to work on all aspects of the common model of an intelligent agent (Sutton, 2022). We commit to building a realization of the OaK architecture laid out in the Alberta Plan (Sutton, Bowling, & Pilarski, 2022).

References

  1. Javed, K., & Sutton, R. S. (2024, July). The big world hypothesis and its ramifications for artificial intelligence. In Finding the Frame: An RLC Workshop for Examining Conceptual Frameworks.
  2. Sutton, R. S. (2022). The quest for a common model of the intelligent decision maker. arXiv preprint arXiv:2202.13252.
  3. Sutton, R. S., Bowling, M., & Pilarski, P. M. (2022). The Alberta plan for AI research. arXiv preprint arXiv:2208.11173.