Deep Symbolic Optimization: Reinforcement Learning for Equation Discovery

Keynote talk on deep symbolic optimization and reinforcement learning for equation discovery at AI for Science 2025. The talk presented DSO as a reinforcement learning approach to symbolic mathematics, in which a neural network learns to generate equations that are rewarded by how well they fit data. It traced the framework’s development from deep symbolic regression to broader applications such as interpretable control policies and generative design in hybrid discrete-continuous spaces.

The event schedule, speaker page, and a survey chapter on the DSO framework are available below: Schedule · Speaker page · Deep Symbolic Optimization: RL for Symbolic Mathematics (Book Chapter)