Poster, Southern California Applied Mathematics Symposium (SoCAMS), Loyola Marymount University, Los Angeles, California
I presented a poster on the semi-algebraic geometry of the space of value functions in reinforcement learning problems at the Southern California Applied Mathematics Symposium (SoCAMS).
Poster, Meetings on Applied Algebraic Geometry (MAAG), Clemson University, Clemson, South Carolina
I presented a poster on the semi-algebraic geometry of the space of value functions in reinforcement learning problems at the Meetings on Applied Algebraic Geometry (MAAG).
Talk, SIAM Conference on Applied Algebraic Geometry (AG25), University of Wisconsin–Madison, Madison, Wisconsin
In this talk I described using interval matrix enclosures to characterize the geometry of the space of value functions in reinforcement learning problems, especially POMDPs.
Talk, UCLA, Department of Statistics, Los Angeles, California
In this talk I described the application of the theory of Grobner bases and elimination to the problem of finding the boundary of the space of value functions in reinforcement learning.