Active Reinforcement Learning for Robust Building Control

TL;DR

This paper introduces ActivePLR, a novel uncertainty-aware environment design algorithm for reinforcement learning that enhances building control by optimizing energy use and occupant comfort under normal and extreme weather conditions.

Contribution

The paper presents ActivePLR, a new UED algorithm that prioritizes performance in a specific environment while improving robustness, specifically applied to building control.

Findings

ActivePLR outperforms existing UED algorithms in energy efficiency.

ActivePLR maintains occupant comfort under extreme weather.

The approach effectively balances performance and robustness.

Abstract

Reinforcement learning (RL) is a powerful tool for optimal control that has found great success in Atari games, the game of Go, robotic control, and building optimization. RL is also very brittle; agents often overfit to their training environment and fail to generalize to new settings. Unsupervised environment design (UED) has been proposed as a solution to this problem, in which the agent trains in environments that have been specially selected to help it learn. Previous UED algorithms focus on trying to train an RL agent that generalizes across a large distribution of environments. This is not necessarily desirable when we wish to prioritize performance in one environment over others. In this work, we will be examining the setting of robust RL building control, where we wish to train an RL agent that prioritizes performing well in normal weather while still being robust to extreme weather conditions. We demonstrate a novel UED algorithm, ActivePLR, that uses uncertainty-aware neural network architectures to generate new training environments at the limit of the RL agent's ability while being able to prioritize performance in a desired base environment. We show that ActivePLR is able to outperform state-of-the-art UED algorithms in minimizing energy usage while maximizing occupant comfort in the setting of building control.