EARCP: Self-Regulating Coherence-Aware Ensemble Architecture for Sequential Decision Making -- Ensemble Auto-Regule par Coherence et Performance

TL;DR

EARCP introduces a dynamic ensemble architecture that adaptively weights models based on performance and coherence, providing theoretical guarantees and robustness in non-stationary sequential decision tasks.

Contribution

It proposes a novel online adaptive ensemble method combining coherence regularization with theoretical regret bounds, applicable across various sequential prediction domains.

Findings

Achieves sublinear regret bounds of O(sqrt(T log M))

Demonstrates effectiveness in time series, activity recognition, and financial prediction

Provides a robust, general-purpose ensemble framework for non-stationary environments

Abstract

We present EARCP (Ensemble Auto-R\'egul\'e par Coh\'erence et Performance), a novel ensemble architecture that dynamically weights heterogeneous expert models based on both their individual performance and inter-model coherence. Unlike traditional ensemble methods that rely on static or offline-learned combinations, EARCP continuously adapts model weights through a principled online learning mechanism that balances exploitation of high-performing models with exploration guided by consensus signals. The architecture combines theoretical foundations from multiplicative weight update algorithms with a novel coherence-based regularization term, providing both theoretical guarantees through regret bounds and practical robustness in non-stationary environments. We formalize the EARCP framework, prove sublinear regret bounds of O(sqrt(T log M)) under standard assumptions, and demonstrate its effectiveness through empirical evaluation on sequential prediction tasks including time series forecasting, activity recognition, and financial prediction. The architecture is designed as a general-purpose framework applicable to any domain requiring ensemble learning with temporal dependencies. An open-source implementation is available at https://github.com/Volgat/earcp and via PyPI (pip install earcp).