In a simplified net-zero microgrid, untuned federated TRPO learns useful battery policies, but tuned PPO gets much closer to the known optimal policy; personal encoding and feature grouping sometimes shrink the gap.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Generalising Battery Control in Net-Zero Buildings via Personalised Federated RL
In a simplified net-zero microgrid, untuned federated TRPO learns useful battery policies, but tuned PPO gets much closer to the known optimal policy; personal encoding and feature grouping sometimes shrink the gap.