Continuous Multi-Objective Reinforcement Learning

Chebyshev Scalarization

A scalarization method using the Chebyshev norm to guarantee the discovery of Pareto-optimal solutions even on non-convex fronts.

← Back