invited-talk

Continuous Verification of Machine Learning: a Declarative Programming Approach

Authors:

Ekaterina Komendantskaya,

Daniel KienitzAuthors Info & Claims

PPDP '20: Proceedings of the 22nd International Symposium on Principles and Practice of Declarative Programming

Article No.: 1, Pages 1 - 3

https://doi.org/10.1145/3414080.3414081

Published: 21 September 2020 Publication History

Abstract

In this invited talk, we discuss state of the art in neural network verification. We propose the term continuous verification to characterise the family of methods that explore continuous nature of machine learning algorithms. We argue that methods of continuous verification must rely on robust programming language infrastructure (refinement types, automated proving, type-driven program synthesis), which provides a major opportunity for the declarative programming language community. Keywords: Neural Networks, Verification, AI.

References

[1]

Behzad Akbarpour and Lawrence Charles Paulson. 2009. MetiTarski: An Automatic Theorem Prover for Real-Valued Special Functions. Journal of Automated Reasoning 44, 3 (Aug. 2009), 175–205. https://doi.org/10.1007/s10817-009-9149-2

[2]

Edward Ayers, Francisco Eiras, Majd Hawasly, and Iain Whiteside. 2020. PaRoT: A Practical Framework for Robust Deep Neural Network Training. In NASA Formal Methods. http://arxiv.org/abs/2001.02152

[3]

Mislav Balunovic and Martin T. Vechev. 2020. Adversarial Training and Provable Defenses: Bridging the Gap. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net.

[4]

Cristiano Calcagno, Dino Distefano, Jérémy Dubreil, Dominik Gabi, Pieter Hooimeijer, Martino Luca, Peter W. O’Hearn, Irene Papakonstantinou, Jim Purbrick, and Dulma Rodriguez. 2015. Moving Fast with Software Verification, In NASA Formal Methods.

[5]

Leonardo Mendonça de Moura and Nikolaj Bjørner. 2008. Z3: An Efficient SMT Solver. In TACAS’08(LNCS), Vol. 4963. 337–340.

[6]

Bruno Dutertre and Leonardo de Moura. 2006. A Fast Linear-Arithmetic Solver for DPLL(T). In Computer Aided Verification. Springer Berlin Heidelberg, 81–94. https://doi.org/10.1007/11817963_11

[7]

Marc Fischer, Mislav Balunovic, Dana Drachsler-Cohen, Timon Gehr, Ce Zhang, and Martin T. Vechev. 2019. DL2: Training and Querying Neural Networks with Logic. In Proc. of the 36th Int. Conf. Machine Learning, ICML 2019, Vol. 97. PMLR, 1931–1941.

[8]

Kathleen Fisher, John Launchbury, and Raymond Richards. 2017. Using Formal Methods to Eliminate Exploitable Bugs. Phil. Trans. R. Soc.(2017).

[9]

Ben Goldberger, Guy Katz, Yossi Adi, and Joseph Keshet. 2020. Minimal Modifications of Deep Neural Networks using Verification. In LPAR 2020: 23rd International Conference on Logic for Programming, Artificial Intelligence and Reasoning, Alicante, Spain, May 22-27, 2020(EPiC Series in Computing), Elvira Albert and Laura Kovács (Eds.), Vol. 73. EasyChair, 260–278.

[10]

Zheng Guo, Michael James, David Justo, Jiaxiao Zhou, Ziteng Wang, Ranjit Jhala, and Nadia Polikarpova. 2020. Program synthesis by type-guided abstraction refinement. Proc. ACM Program. Lang. 4, POPL (2020), 12:1–12:28.

Digital Library

[11]

Xiaowei Huang, Marta Kwiatkowska, Sen Wang, and Min Wu. 2017. Safety Verification of Deep Neural Networks. In CAV 2017, Vol. LNCS 10426. Springer, 3–29. https://doi.org/10.1007/978-3-319-63387-9

[12]

Simon Peyton Jones, Stephanie Weirich, Richard A. Eisenberg, and Dimitrios Vytiniotis. 2016. A Reflection on Types — A List of Successes That Can Change the World. (2016).

[13]

Dejan Jovanović and Leonardo de Moura. 2013. Solving non-linear arithmetic. ACM Communications in Computer Algebra 46, 3/4 (Jan. 2013), 104. https://doi.org/10.1145/2429135.2429155

Digital Library

[14]

Guy Katz, Derek A. Huang, Duligur Ibeling, Kyle Julian, Christopher Lazarus, Rachel Lim, Parth Shah, Shantanu Thakoor, Haoze Wu, Aleksandar Zeljic, David L. Dill, Mykel J. Kochenderfer, and Clark W. Barrett. 2019. The Marabou Framework for Verification and Analysis of Deep Neural Networks. In CAV 2019, Part I(LNCS), Vol. 11561. Springer, 443–452.

[15]

Tristan Knoth, Di Wang, Adam Reynolds, Jan Hoffmann, and Nadia Polikarpova. 2020. Liquid resource types. Proc. ACM Program. Lang. 4, ICFP (2020), 106:1–106:29.

Digital Library

[16]

Wen Kokke, Ekaterina Komendantskaya, and Daniel Kienitz. 2020. StarChild, a library for leveraging the refinement types and SMT solving of F* to verify properties of neural networks. https://github.com/wenkokke/starchild

[17]

Wen Kokke, Ekaterina Komendantskaya, Daniel Kienitz, Robert Atkey, and David Aspinall. 2020. Neural Networks, Secure by Construction: An Exploration of Refinement Types. Draft.

[18]

Yann LeCun, Corinna Cortes, and CJ Burges. 2010. MNIST handwritten digit database. ATT Labs [Online]. Available: http://yann.lecun.com/exdb/mnist 2(2010).

[19]

Steve McConnell. 2015. Code Complete. A Practical Handbook of Software Construction, Second Edition.

[20]

Gagandeep Singh, Timon Gehr, Markus Püschel, and Martin T. Vechev. 2019. An abstract domain for certifying neural networks. PACMPL 3, POPL (2019), 41:1–41:30. https://doi.org/10.1145/3290354

[21]

Nikhil Swamy, Markulf Kohlweiss, Jean-Karim Zinzindohoue, Santiago Zanella-Béguelin, Cătălin Hriţcu, Chantal Keller, Aseem Rastogi, Antoine Delignat-Lavaud, Simon Forest, Karthikeyan Bhargavan, Cédric Fournet, and Pierre-Yves Strub. 2016. Dependent types and multi-monadic effects in F*. In Proceedings of the 43rd Annual ACM SIGPLAN-SIGACT Symposium on Principles of Programming Languages - POPL 2016. ACM Press. https://doi.org/10.1145/2837614.2837655

Digital Library

[22]

Christian Szegedy, Wojciech Zaremba, Ilya Sutskever, Joan Bruna, Dumitru Erhan, Ian Goodfellow, and Rob Fergus. 2014. Intriguing properties of neural networks. CoRR abs/1312.6199(2014). https://arxiv.org/abs/1312.6199

[23]

Niki Vazou. 2016. Liquid Haskell: Haskell as a Theorem Prover. Ph.D. Dissertation. University of California, San Diego, USA.

Cited By

Ślusarz NKomendantskaya EDaggitt MStewart R(2022)Differentiable Logics for Neural Network Training and VerificationSoftware Verification and Formal Methods for ML-Enabled Autonomous Systems10.1007/978-3-031-21222-2_5(67-77)Online publication date: 16-Dec-2022
https://doi.org/10.1007/978-3-031-21222-2_5
Casadio MKomendantskaya EDaggitt MKokke WKatz GAmir GRefaeli I(2022)Neural Network Robustness as a Verification Property: A Principled Case StudyComputer Aided Verification10.1007/978-3-031-13185-1_11(219-231)Online publication date: 7-Aug-2022
https://dl.acm.org/doi/10.1007/978-3-031-13185-1_11

Continuous Verification of Machine Learning: a Declarative Programming Approach
1. Computing methodologies
  1. Machine learning
    1. Machine learning approaches

Recommendations

A Robust Approach for Continuous Interactive Reinforcement Learning
HAI '20: Proceedings of the 8th International Conference on Human-Agent Interaction

Interactive reinforcement learning is an approach in which an external trainer helps an agent to learn through advice. A trainer is useful in large or continuous scenarios; however, when the characteristics of the environment change over time, it can ...
Continuous Verification by Discrete Reasoning
Abductive Logic Programming as an Effective Technology for the Static Verification of Declarative Business Processes
RCRA 2008 Experimental Evaluation of Algorithms for Solving Problems with Combinatorial Explosion

We discuss the static verification of declarative Business Processes. We identify four desiderata about verifiers, and propose a concrete framework which satisfies them. The framework is based on the ConDec graphical notation for modeling Business ...

Comments

Information & Contributors

Information

Published In

cover image ACM Other conferences

PPDP '20: Proceedings of the 22nd International Symposium on Principles and Practice of Declarative Programming

September 2020

179 pages

ISBN:9781450388214

DOI:10.1145/3414080

General Chair:
Maurizio Gabbrielli
Department of Computer Science and Engineering, University of Bologna, Italy
,
Program Chairs:
Andreas Abel
Gothenburg University, Sweden
,
James Cheney
The University of Edinburgh, UK

Copyright © 2020 Owner/Author.

Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for third-party components of this work must be honored. For all other uses, contact the Owner/Author.

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 21 September 2020

Check for updates

Qualifiers

Invited-talk
Research
Refereed limited

Conference

PPDP '20

PPDP '20: 22nd International Symposium on Principles and Practice of Declarative Programming

September 8 - 10, 2020

Bologna, Italy

Acceptance Rates

Overall Acceptance Rate 230 of 486 submissions, 47%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

2
Total Citations
View Citations
130
Total Downloads

Downloads (Last 12 months)18
Downloads (Last 6 weeks)0

Reflects downloads up to 22 Sep 2024

Other Metrics

View Author Metrics

Citations

Cited By

Ślusarz NKomendantskaya EDaggitt MStewart R(2022)Differentiable Logics for Neural Network Training and VerificationSoftware Verification and Formal Methods for ML-Enabled Autonomous Systems10.1007/978-3-031-21222-2_5(67-77)Online publication date: 16-Dec-2022
https://doi.org/10.1007/978-3-031-21222-2_5
Casadio MKomendantskaya EDaggitt MKokke WKatz GAmir GRefaeli I(2022)Neural Network Robustness as a Verification Property: A Principled Case StudyComputer Aided Verification10.1007/978-3-031-13185-1_11(219-231)Online publication date: 7-Aug-2022
https://dl.acm.org/doi/10.1007/978-3-031-13185-1_11

View Options

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

HTML Format

View this article in HTML Format.

Media

Figures

Other

Tables

View Table of Contents