Neuro-Evolved Heuristics for Variable Gapped Common Subsequence Identification

2026-08-01

AI Summary

This work addresses the challenge of insufficient robustness in existing heuristic methods for the variable-gap constrained longest common subsequence problem (VGLCSP) when applied to multiple sequences. To overcome this limitation, we propose a neural-evolution-based heuristic learning framework that, for the first time, integrates neural evolution into VGLCSP. Our approach employs a fixed-architecture neural network to automatically learn data-driven heuristic functions within an iterative multi-source beam search procedure and combines these learned heuristics with the best handcrafted ones to form a hybrid strategy, thereby enhancing both generalization and solution quality. Experimental results demonstrate that the proposed method significantly outperforms current state-of-the-art techniques on both synthetic benchmarks and newly constructed real-world instances, confirming its effectiveness and superiority.

Abstract

This study addresses the Variable Gapped Longest Common Subsequence Problem (VGLCSP), a variant of the classical longest common subsequence problem with additional gap constraints and applications in sequence alignment and time-series analysis. While the two-sequence version has been widely studied using dynamic programming, the generalized multi-sequence form is usually solved with beam search-based heuristics, whose hand-crafted designs often lack robustness. To overcome this limitation, we propose a learning-based approach for automatically designing more effective data-driven heuristics. The heuristics are represented by a neural network with predefined architecture, whose weights are optimized by a genetic algorithm within a neuro-evolutionary framework. The learning process alternates between weight optimization and evaluation within an iterative multi-source beam search procedure, a state-of-the-art method for the problem. Rather than constructing solutions directly, the neural network learns to guide the search process, producing a neuro-evolved heuristic. We further introduce an ensemble heuristic that combines the scores of learned and the best-performing hand-crafted heuristic. Integrated into the iterative multi-source beam search framework, the resulting hybrid approach outperforms existing methods on both synthetic benchmark instances and newly introduced real-world instances with data-driven gap constraints.

Problem

• Variable Gapped Longest Common Subsequence
• multi-sequence alignment
• heuristic robustness
• sequence analysis
• gap constraints

Innovation

• neuro-evolution
• learned heuristics
• variable gapped subsequence
• beam search
• ensemble heuristic