Finite-Memory Strategies for Almost-Sure Energy-MeanPayoff Objectives in MDPs.

Mohan Dantam Richard Mayr

Published in: ICALP (2024)

Keyphrases

markov decision processes
memory requirements
reinforcement learning
state and action spaces
energy consumption
energy minimization
state space
finite number
memory space
neural network
finite state
optimal strategy
memory usage
random access
average cost
markov decision process
finite horizon