Meta-Learning-Based Deep Reinforcement Learning for Multiobjective Optimization Problems
This work addresses multiobjective combinatorial optimization problems, such as traveling salesman and vehicle routing, offering a more flexible and efficient method compared to existing approaches, though it is incremental in nature.
The paper tackles the challenge of efficiently solving multiobjective optimization problems by proposing a meta-learning-based deep reinforcement learning approach that trains a meta-model and fine-tunes it for subproblems, resulting in greatly shortened training times and improved solution quality and diversity.
Deep reinforcement learning (DRL) has recently shown its success in tackling complex combinatorial optimization problems. When these problems are extended to multiobjective ones, it becomes difficult for the existing DRL approaches to flexibly and efficiently deal with multiple subproblems determined by weight decomposition of objectives. This paper proposes a concise meta-learning-based DRL approach. It first trains a meta-model by meta-learning. The meta-model is fine-tuned with a few update steps to derive submodels for the corresponding subproblems. The Pareto front is then built accordingly. Compared with other learning-based methods, our method can greatly shorten the training time of multiple submodels. Due to the rapid and excellent adaptability of the meta-model, more submodels can be derived so as to increase the quality and diversity of the found solutions. The computational experiments on multiobjective traveling salesman problems and multiobjective vehicle routing problem with time windows demonstrate the superiority of our method over most of learning-based and iteration-based approaches.