The Computational Minds and Agent Well-Being (CMAW) Lab develops the mathematical foundations of computational theories of mind and their implications for AI. Our research addresses two questions.

Foundations: First, how can we mathematically characterize the relationship between computation, physical systems, and agent experience?

Qualia Optimization: Second, given such a characterization, how can we design AI systems that achieve their intended goals while also optimizing for agent well-being?

We combine theoretical analysis with empirical algorithm evaluation, drawing on machine learning, reinforcement learning, and theoretical computer science.