Prediction horizon vs. efficiency of optimal dynamic thermal control policies in HPC nodes.

Daniele Cesarini Andrea Bartolini Luca Benini

Published in: VLSI-SoC (2017)

Keyphrases

control policies
control policy
reinforcement learning
optimal policy
dynamic environments
control strategies
motion control
stochastic optimization problems
control system
high performance computing
action space
optimal solution
neural network
long run
shortest path
finite horizon
dynamic programming