Efficient Neural Hybrid System Learning and Transition System Abstraction for Dynamical Systems

TL;DR

This paper introduces a neural hybrid modeling framework that combines local neural networks and high-level transition system abstraction to improve interpretability, efficiency, and verifiability in dynamical systems learning.

Contribution

It presents a novel two-level neural hybrid modeling approach that integrates data-driven local dynamics learning with transition system abstraction for enhanced verification.

Findings

Efficient dynamics learning through neural network partitions

High-level transition system enables formal verification

Improved interpretability and computational efficiency

Abstract

This paper proposes a neural network hybrid modeling framework for dynamics learning to promote an interpretable, computationally efficient way of dynamics learning and system identification. First, a low-level model will be trained to learn the system dynamics, which utilizes multiple simple neural networks to approximate the local dynamics generated from data-driven partitions. Then, based on the low-level model, a high-level model will be trained to abstract the low-level neural hybrid system model into a transition system that allows Computational Tree Logic Verification to promote the model's ability with human interaction and verification efficiency.