Gradient-Based Neuroplastic Adaptation for Concurrent Optimization of Neuro-Fuzzy Networks

TL;DR

This paper introduces a gradient-based neuroplastic adaptation method that enables simultaneous optimization of neuro-fuzzy networks' parameters and structure, improving their applicability in online reinforcement learning for vision-based tasks.

Contribution

It presents a novel, application-independent approach for concurrent optimization of NFNs' parameters and structure, addressing limitations of sequential design methods.

Findings

Successfully applied to online reinforcement learning in a vision-based video game.

Demonstrated improved adaptability and performance of NFNs.

Enabled new applications previously inaccessible due to design constraints.

Abstract

Neuro-fuzzy networks (NFNs) are transparent, symbolic, and universal function approximations that perform as well as conventional neural architectures, but their knowledge is expressed as linguistic IF-THEN rules. Despite these advantages, their systematic design process remains a challenge. Existing work will often sequentially build NFNs by inefficiently isolating parametric and structural identification, leading to a premature commitment to brittle and subpar architecture. We propose a novel application-independent approach called gradient-based neuroplastic adaptation for the concurrent optimization of NFNs' parameters and structure. By recognizing that NFNs' parameters and structure should be optimized simultaneously as they are deeply conjoined, settings previously unapproachable for NFNs are now accessible, such as the online reinforcement learning of NFNs for vision-based tasks. The effectiveness of concurrently optimizing NFNs is empirically shown as it is trained by online reinforcement learning to proficiently play challenging scenarios from a vision-based video game called DOOM.