PolyGym: Polyhedral Optimizations as an Environment for Reinforcement Learning.

Alexander Brauckmann Andrés Goens Jerónimo Castrillón

Published in: PACT (2021)

Keyphrases

reinforcement learning
real time
learning algorithm
multi agent environments
supervised learning
convex hull
function approximation
neural network
multi agent
learning process
dynamic programming
agent receives