Multi-Objective Optimization Using Adaptive Distributed Reinforcement Learning
This addresses the challenge of multi-objective optimization in distributed ITS environments, where existing single-objective RL methods are insufficient, though it is incremental as it builds on MARL techniques.
The paper tackles the problem of optimizing multiple conflicting objectives in dynamic Intelligent Transportation Systems (ITS) by proposing a multi-objective, multi-agent reinforcement learning algorithm, which shows quick adaptation and outperforms state-of-the-art benchmarks in all metrics, with inference times as low as 6 milliseconds on a single-board computer.
The Intelligent Transportation System (ITS) environment is known to be dynamic and distributed, where participants (vehicle users, operators, etc.) have multiple, changing and possibly conflicting objectives. Although Reinforcement Learning (RL) algorithms are commonly applied to optimize ITS applications such as resource management and offloading, most RL algorithms focus on single objectives. In many situations, converting a multi-objective problem into a single-objective one is impossible, intractable or insufficient, making such RL algorithms inapplicable. We propose a multi-objective, multi-agent reinforcement learning (MARL) algorithm with high learning efficiency and low computational requirements, which automatically triggers adaptive few-shot learning in a dynamic, distributed and noisy environment with sparse and delayed reward. We test our algorithm in an ITS environment with edge cloud computing. Empirical results show that the algorithm is quick to adapt to new environments and performs better in all individual and system metrics compared to the state-of-the-art benchmark. Our algorithm also addresses various practical concerns with its modularized and asynchronous online training method. In addition to the cloud simulation, we test our algorithm on a single-board computer and show that it can make inference in 6 milliseconds.