Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Yejiang Yang, Zihao Mo, Weiming Xiang

TL;DR

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.

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.

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

TL;DR

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.

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.

Paper Structure

This paper contains 12 sections, 17 equations, 1 figure, 3 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Notations.
  3. Preliminaries and Problem Formulation
  4. Reachability Analysis for Neural Network Dynamical System
  5. Maximum Entropy Partitioning
  6. Problem Formulation
  7. Dual-Level Modeling Framework
  8. Neural Hybrid System Model and Transition System Abstraction
  9. Efficient Dynamics Learning via Low-Level Modeling
  10. Interpretable Abstraction via High-Level Model
  11. Applications to Dynamical System Modeling
  12. Conclusion

Figures (1)

  • Figure 1: Partitions, Cells, and Transition Map Abstraction of Dual-Level Models for $MultiModels_2$ from LASA data set.

Theorems & Definitions (4)

  • Definition 1
  • Definition 2
  • Definition 3
  • Remark 1