Reinforcement-guided hyper-heuristic hyperparameter optimization for fair and explainable spiking neural network-based financial fraud detection

Sadman Mohammad Nasif

doi:10.1016/j.knosys.2026.115814

Reinforcement-guided hyper-heuristic hyperparameter optimization for fair and explainable spiking neural network-based financial fraud detection

2026-03-24

0

PRE

AI

S

Sadman Mohammad Nasif

DOI：10.1016/j.knosys.2026.115814

Original

Original request for help

Save

Abstract

En 中文

• RHOSS uses Q-learning to guide fairness-aware SNN hyperparameter optimization. • CSNPC improves recall at 5% FPR using structured population-coded spiking outputs. • MoSSTI framework enables dual-path explainability via saliency and spike profiling. • Achieves 90.8% recall and >98% fairness on real-world financial fraud benchmarks. • Outperforms classical and SNN baselines under fairness and transparency constraints.

Keywords:

RHOSS

CSNPC

MoSSTI

fairness-aware SNN

explainable spiking neural networks

Reinforcement-guided hyper-heuristic hyperparameter optimization for fair and explainable spiking neural network-based financial fraud detection

Abstract

Journal

Organization