Data-Driven Abstractions via Binary-Tree Gaussian Processes for Formal Verification

被引:0
|
作者
Schon, Oliver [1 ]
Naseer, Shammakh [1 ]
Wooding, Ben [1 ]
Soudjani, Sadegh [2 ]
机构
[1] Newcastle Univ, Sch Comp, Newcastle Upon Tyne, Tyne & Wear, England
[2] Max Planck Inst Software Syst, Saarbrucken, Germany
来源
IFAC PAPERSONLINE | 2024年 / 58卷 / 11期
关键词
Stochastic Systems; Formal Verification; Gaussian Processes; Uncertain Systems; Error Quantification; Machine Learning; System Identification;
D O I
10.1016/j.ifacol.2024.07.434
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
To advance formal verification of stochastic systems against temporal logic requirements for handling unknown dynamics, researchers have been designing data-driven approaches inspired by breakthroughs in the underlying machine learning techniques. As one promising research direction, abstraction-based solutions based on Gaussian process (GP) regression have become popular for their ability to learn a representation of the latent system from data with a quantified error. Results obtained based on this model are then translated to the true system via various methods. In a recent publication, GPs using a so-called binary-tree kernel have demonstrated a polynomial speedup w.r.t. the size of the data compared to their vanilla version, outcompeting all existing sparse GP approximations. Incidentally, the resulting binary-tree Gaussian process (BTGP) is characteristic for its piecewise-constant posterior mean and covariance functions, naturally abstracting the input space into discrete partitions. In this paper, we leverage this natural abstraction of the BTGP for formal verification, eliminating the need for cumbersome abstraction and error quantification procedures. We show that the BTGP allows us to construct an interval Markov chain model of the unknown system with a speedup that is polynomial w.r.t. the size of the abstraction compared to alternative approaches. We provide a delocalized error quantification via a unified formula even when the true dynamics do not live in the function space of the BTGP. This allows us to compute upper and lower bounds on the probability of satisfying reachability specifications that are robust to both aleatoric and epistemic uncertainties. Copyright (c) 2024 The Authors.
引用
收藏
页码:115 / 122
页数:8
相关论文
共 50 条
  • [21] Formal verification of complex systems: model-based and data-driven methods
    Abate, Alessandro
    MEMOCODE 2017: PROCEEDINGS OF THE 15TH ACM-IEEE INTERNATIONAL CONFERENCE ON FORMAL METHODS AND MODELS FOR SYSTEM DESIGN, 2017, : 92 - 94
  • [22] Linearized Gaussian Processes for Fast Data-driven Model Predictive Control
    Nghiem, Truong X.
    2019 AMERICAN CONTROL CONFERENCE (ACC), 2019, : 1629 - 1634
  • [23] Data-driven Demand Response Modeling and Control of Buildings with Gaussian Processes
    Nghiem, Truong X.
    Jones, Colin N.
    2017 AMERICAN CONTROL CONFERENCE (ACC), 2017, : 2919 - 2924
  • [24] Spatial Modeling of Precipitation Based on Data-Driven Warping of Gaussian Processes
    Agou, Vasiliki D.
    Pavlides, Andrew
    Hristopulos, Dionissios T.
    ENTROPY, 2022, 24 (03)
  • [25] Stochastic data-driven model predictive control using gaussian processes
    Bradford, Eric
    Imsland, Lars
    Zhang, Dongda
    Chanona, Ehecatl Antonio del Rio
    COMPUTERS & CHEMICAL ENGINEERING, 2020, 139
  • [26] Data-Driven Synthesis of Symbolic Abstractions With Guaranteed Confidence
    Lavaei, Abolfazl
    Frazzoli, Emilio
    IEEE CONTROL SYSTEMS LETTERS, 2022, 7 : 253 - 258
  • [27] Scalable Gaussian Processes for Data-Driven Design Using Big Data With Categorical Factors
    Wang, Liwei
    Yerramilli, Suraj
    Iyer, Akshay
    Apley, Daniel
    Zhu, Ping
    Chen, Wei
    JOURNAL OF MECHANICAL DESIGN, 2022, 144 (02)
  • [28] Globally Approximate Gaussian Processes for Big Data With Application to Data-Driven Metamaterials Design
    Bostanabad, Ramin
    Chan, Yu-Chin
    Wang, Liwei
    Zhu, Ping
    Chen, Wei
    JOURNAL OF MECHANICAL DESIGN, 2019, 141 (11)
  • [29] Data-driven inference of passivity properties via Gaussian process optimization
    Romer, Anne
    Trimpe, Sebastian
    Allgoewer, Frank
    2019 18TH EUROPEAN CONTROL CONFERENCE (ECC), 2019, : 29 - 35
  • [30] Data-driven verification and synthesis of stochastic systems via barrier certificates
    Salamati, Ali
    Lavaei, Abolfazl
    Soudjani, Sadegh
    Zamani, Majid
    AUTOMATICA, 2024, 159