COBRA++: Enhanced COBRA Optimizer with Augmented Surrogate Pool and Reinforced Surrogate Selection

TL;DR

COBRA++ introduces an adaptive, learning-based enhancement to the COBRA optimizer by augmenting its surrogate pool and employing reinforcement learning for model selection, significantly improving optimization performance on complex problems.

Contribution

The paper presents a novel, learning-based adaptive strategy for COBRA optimizer, automating surrogate model selection and diversifying the surrogate pool to improve optimization efficiency.

Findings

COBRA++ outperforms vanilla COBRA and its adaptive variant in diverse optimization tasks.

The reinforcement learning-based model selection improves optimization accuracy.

Ablation studies confirm the effectiveness of each design component.

Abstract

The optimization problems in realistic world present significant challenges onto optimization algorithms, such as the expensive evaluation issue and complex constraint conditions. COBRA optimizer (including its up-to-date variants) is a representative and effective tool for addressing such optimization problems, which introduces 1) RBF surrogate to reduce online evaluation and 2) bi-stage optimization process to alternate search for feasible solution and optimal solution. Though promising, its design space, i.e., surrogate model pool and selection standard, is still manually decided by human expert, resulting in labor-intensive fine-tuning for novel tasks. In this paper, we propose a learning-based adaptive strategy (COBRA++) that enhances COBRA in two aspects: 1) An augmented surrogate pool to break the tie with RBF-like surrogate and hence enhances model diversity and approximation capability; 2) A reinforcement learning-based online model selection policy that empowers efficient and accurate optimization process. The model selection policy is trained to maximize overall performance of COBRA++ across a distribution of constrained optimization problems with diverse properties. We have conducted multi-dimensional validation experiments and demonstrate that COBRA++ achieves substantial performance improvement against vanilla COBRA and its adaptive variant. Ablation studies are provided to support correctness of each design component in COBRA++.