Explaining deep learning models - A Bayesian non-parametric approach

Wenbo Guo, Sui Huang, Yunzhe Tao, Xinyu Xing, Lin Lin

Research output: Contribution to journalConference article

Abstract

Understanding and interpreting how machine learning (ML) models make decisions have been a big challenge. While recent research has proposed various technical approaches to provide some clues as to how an ML model makes individual predictions, they cannot provide users with an ability to inspect a model as a complete entity. In this work, we propose a novel technical approach that augments a Bayesian non-parametric regression mixture model with multiple elastic nets. Using the enhanced mixture model, we can extract generalizable insights for a target model through a global approximation. To demonstrate the utility of our approach, we evaluate it on different ML models in the context of image recognition. The empirical results indicate that our proposed approach not only outperforms the state-of-the-art techniques in explaining individual decisions but also provides users with an ability to discover the vulnerabilities of the target ML models.

Original languageEnglish (US)
Pages (from-to)4514-4524
Number of pages11
JournalAdvances in Neural Information Processing Systems
Volume2018-December
StatePublished - Jan 1 2018
Event32nd Conference on Neural Information Processing Systems, NeurIPS 2018 - Montreal, Canada
Duration: Dec 2 2018Dec 8 2018

Fingerprint

Learning systems
Image recognition
Deep learning

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Information Systems
  • Signal Processing

Cite this

@article{575b36ef30de4eb1969ec3844f82963f,
title = "Explaining deep learning models - A Bayesian non-parametric approach",
abstract = "Understanding and interpreting how machine learning (ML) models make decisions have been a big challenge. While recent research has proposed various technical approaches to provide some clues as to how an ML model makes individual predictions, they cannot provide users with an ability to inspect a model as a complete entity. In this work, we propose a novel technical approach that augments a Bayesian non-parametric regression mixture model with multiple elastic nets. Using the enhanced mixture model, we can extract generalizable insights for a target model through a global approximation. To demonstrate the utility of our approach, we evaluate it on different ML models in the context of image recognition. The empirical results indicate that our proposed approach not only outperforms the state-of-the-art techniques in explaining individual decisions but also provides users with an ability to discover the vulnerabilities of the target ML models.",
author = "Wenbo Guo and Sui Huang and Yunzhe Tao and Xinyu Xing and Lin Lin",
year = "2018",
month = "1",
day = "1",
language = "English (US)",
volume = "2018-December",
pages = "4514--4524",
journal = "Advances in Neural Information Processing Systems",
issn = "1049-5258",

}

Explaining deep learning models - A Bayesian non-parametric approach. / Guo, Wenbo; Huang, Sui; Tao, Yunzhe; Xing, Xinyu; Lin, Lin.

In: Advances in Neural Information Processing Systems, Vol. 2018-December, 01.01.2018, p. 4514-4524.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Explaining deep learning models - A Bayesian non-parametric approach

AU - Guo, Wenbo

AU - Huang, Sui

AU - Tao, Yunzhe

AU - Xing, Xinyu

AU - Lin, Lin

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Understanding and interpreting how machine learning (ML) models make decisions have been a big challenge. While recent research has proposed various technical approaches to provide some clues as to how an ML model makes individual predictions, they cannot provide users with an ability to inspect a model as a complete entity. In this work, we propose a novel technical approach that augments a Bayesian non-parametric regression mixture model with multiple elastic nets. Using the enhanced mixture model, we can extract generalizable insights for a target model through a global approximation. To demonstrate the utility of our approach, we evaluate it on different ML models in the context of image recognition. The empirical results indicate that our proposed approach not only outperforms the state-of-the-art techniques in explaining individual decisions but also provides users with an ability to discover the vulnerabilities of the target ML models.

AB - Understanding and interpreting how machine learning (ML) models make decisions have been a big challenge. While recent research has proposed various technical approaches to provide some clues as to how an ML model makes individual predictions, they cannot provide users with an ability to inspect a model as a complete entity. In this work, we propose a novel technical approach that augments a Bayesian non-parametric regression mixture model with multiple elastic nets. Using the enhanced mixture model, we can extract generalizable insights for a target model through a global approximation. To demonstrate the utility of our approach, we evaluate it on different ML models in the context of image recognition. The empirical results indicate that our proposed approach not only outperforms the state-of-the-art techniques in explaining individual decisions but also provides users with an ability to discover the vulnerabilities of the target ML models.

UR - http://www.scopus.com/inward/record.url?scp=85064814680&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064814680&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85064814680

VL - 2018-December

SP - 4514

EP - 4524

JO - Advances in Neural Information Processing Systems

JF - Advances in Neural Information Processing Systems

SN - 1049-5258

ER -