Optimisation problems, particularly combinatorial optimisation problems, are
difficult to solve due to their complexity and hardness. Such problems have
been successfully solved by evolutionary and swarm intelligence algorithms,
especially in binary format. However, the approximation may suffer due to the
the issues in balance between exploration and exploitation activities (EvE),
which remain as the major challenge in this context. Although the complementary
usage of multiple operators is becoming more popular for managing EvE with
adaptive operator selection schemes, a bespoke adaptive selection system is
still an important topic in research. Reinforcement Learning (RL) has recently
been proposed as a way to customise and shape up a highly effective adaptive
selection system. However, it is still challenging to handle the problem in
terms of scalability. This paper proposes and assesses a RL-based novel
approach to help develop a generalised framework for gaining, processing, and
utilising the experiences for both the immediate and future use. The
experimental results support the proposed approach with a certain level of
success.
Optimisation problems, especially combinatorial ones, are notoriously difficult to solve due to their complexity and hardness. Evolutionary and swarm intelligence algorithms have shown promise in solving such problems, particularly in binary format. However, these algorithms face a major challenge in balancing exploration and exploitation activities, which can affect the quality of the approximation.
To address this challenge, researchers have started considering the use of multiple operators in combination with adaptive operator selection schemes. While this approach has gained popularity, there is still a need for a bespoke adaptive selection system that can effectively manage exploration and exploitation.
One potential solution that has emerged is Reinforcement Learning (RL). RL is a learning approach that involves an agent interacting with an environment and learning through trial and error. This paper proposes leveraging RL to develop a highly effective adaptive selection system for optimisation problems. By customizing and shaping the selection process using RL, the researchers aim to overcome the challenges associated with exploration and exploitation.
However, scalability remains a concern when applying RL to optimisation problems. As the complexity of the problems increases, RL algorithms can struggle to handle the large state and action spaces effectively.
The proposed approach in this paper aims to address this scalability issue by developing a generalised framework for gaining, processing, and utilising experiences. The framework takes into account both immediate and future use of experiences, allowing for more robust decision-making.
The experimental results presented in this paper provide evidence of the success of the proposed RL-based approach. However, further research is required to assess its performance across a wider range of optimisation problems and to explore potential refinements to enhance scalability.
Key Takeaways:
- Optimisation problems, especially combinatorial ones, are challenging due to their complexity and hardness.
- Evolutionary and swarm intelligence algorithms have shown promise in solving these problems, but the balance between exploration and exploitation remains a challenge.
- Reinforcement Learning (RL) has been proposed as a way to develop effective adaptive selection systems, but scalability is an issue.
- This paper proposes an RL-based approach to overcome scalability challenges and develop a generalised framework for optimisation problems.
- Experimental results support the proposed approach, but further research is needed to validate its performance on a broader range of problems.