SMT-Based Translation Validation for Machine Learning Compiler

被引:7
|
作者
Bang, Seongwon [1 ]
Nam, Seunghyeon [1 ]
Chun, Inwhan [1 ]
Jhoo, Ho Young [1 ]
Lee, Juneyoung [2 ]
机构
[1] Seoul Natl Univ, Seoul, South Korea
[2] CryptoLab, Seoul, South Korea
基金
新加坡国家研究基金会;
关键词
D O I
10.1007/978-3-031-13188-2_19
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Machine learning compilers are large software containing complex transformations for deep learning models, and any buggy transformation may cause a crash or silently bring a regression to the prediction accuracy and performance. This paper proposes an SMT-based translation validation framework for Multi-Level IR (MLIR), a compiler framework used by many deep learning compilers. It proposes an SMT encoding tailored for translation validation that is an over-approximation of the FP arithmetic and reduction operations. It performs abstraction refinement if validation fails. We also propose a new approach for encoding arithmetic properties of reductions in SMT. We found mismatches between the specification and implementation of MLIR, and validated high-level transformations for SqueezeNet, MobileNet, and text_classification with proper splitting.
引用
收藏
页码:386 / 407
页数:22
相关论文
共 50 条
  • [31] Practical SMT-Based Type Error Localization
    Pavlinovic, Zvonimir
    King, Tim
    Wies, Thomas
    ACM SIGPLAN NOTICES, 2015, 50 (09) : 412 - 423
  • [32] SMT-Based Aircraft Conflict Detection and Resolution
    Paul, Saswata
    Meng, Baoluo
    Alexander, Christopher
    NASA FORMAL METHODS, NFM 2024, 2024, 14627 : 186 - 203
  • [33] SMT-based scenario verification for hybrid systems
    Cimatti, Alessandro
    Mover, Sergio
    Tonetta, Stefano
    FORMAL METHODS IN SYSTEM DESIGN, 2013, 42 (01) : 46 - 66
  • [34] Invariant Checking for SMT-Based Systems with Quantifiers
    Redondi, Gianluca
    Cimatti, Alessandro
    Griggio, Alberto
    Mcmillan, Kenneth L.
    ACM TRANSACTIONS ON COMPUTATIONAL LOGIC, 2024, 25 (04)
  • [35] Scalable Lazy SMT-Based Motion Planning
    Shoukry, Yasser
    Nuzzo, Pierluigi
    Saha, Indranil
    Sangiovanni-Vincentelli, Alberto L.
    Seshia, Sanjit A.
    Pappas, George J.
    Tabuada, Paulo
    2016 IEEE 55TH CONFERENCE ON DECISION AND CONTROL (CDC), 2016, : 6683 - 6688
  • [36] SMT-Based Model Checking for Recursive Programs
    Komuravelli, Anvesh
    Gurfinkel, Arie
    Chaki, Sagar
    COMPUTER AIDED VERIFICATION, CAV 2014, 2014, 8559 : 17 - 34
  • [37] SMT-Based Encoding of Argumentation Dialogue Games
    Kacprzak, Magdalena
    Sawicka, Anna
    Zbrzezny, Andrzej
    ARTIFICIAL INTELLIGENCE AND SOFT COMPUTING, ICAISC 2019, PT II, 2019, 11509 : 564 - 574
  • [38] A Unifying View on SMT-Based Software Verification
    Dirk Beyer
    Matthias Dangl
    Philipp Wendler
    Journal of Automated Reasoning, 2018, 60 : 299 - 335
  • [39] Evaluation of Cardinality Constraints on SMT-based Debugging
    Suelflow, Andre
    Wille, Robert
    Fey, Goerschwin
    Drechsler, Rolf
    ISMVL: 2009 39TH IEEE INTERNATIONAL SYMPOSIUM ON MULTIPLE-VALUED LOGIC, 2009, : 298 - 303
  • [40] A Unifying View on SMT-Based Software Verification
    Beyer, Dirk
    Dangl, Matthias
    Wendler, Philipp
    JOURNAL OF AUTOMATED REASONING, 2018, 60 (03) : 299 - 335