Open Internet by MindsNet

Solving Constrained Optimization Problems for Safe Reinforcement Learning

The author is struggling to solve a constrained optimization problem using the augmented Lagrangian method for a safe reinforcement learning problem in robotics. They are using control Lyapunov and Barrier functions to form constraints but are having trouble proceeding. This indicates a need for better understanding or tools for solving such optimization problems.

Mathematics & logic, Mathematics, Optimization