skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Interpreting Black-Box Classifiers Using Instance-Level Visual Explanations

Abstract

To realize the full potential of machine learning in diverse real- world domains, it is necessary for model predictions to be readily interpretable and actionable for the human in the loop. Analysts, who are the users but not the developers of machine learning models, often do not trust a model because of the lack of transparency in associating predictions with the underlying data space. To address this problem, we propose Rivelo, a visual analytic interface that enables analysts to understand the causes behind predictions of binary classifiers by interactively exploring a set of instance-level explanations. These explanations are model-agnostic, treating a model as a black box, and they help analysts in interactively probing the high-dimensional binary data space for detecting features relevant to predictions. We demonstrate the utility of the interface with a case study analyzing a random forest model on the sentiment of Yelp reviews about doctors.

Authors:
; ; ;
Publication Date:
Research Org.:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1358512
Report Number(s):
PNNL-SA-124676
DOE Contract Number:
AC05-76RL01830
Resource Type:
Conference
Resource Relation:
Conference: Proceedings of the 2nd Workshop on Human-in-the-Loop Data Analytics (HILDA 2017) May 14-19, 2017, Chicago, Illinois, Article No. 6
Country of Publication:
United States
Language:
English

Citation Formats

Tamagnini, Paolo, Krause, Josua W., Dasgupta, Aritra, and Bertini, Enrico. Interpreting Black-Box Classifiers Using Instance-Level Visual Explanations. United States: N. p., 2017. Web. doi:10.1145/3077257.3077260.
Tamagnini, Paolo, Krause, Josua W., Dasgupta, Aritra, & Bertini, Enrico. Interpreting Black-Box Classifiers Using Instance-Level Visual Explanations. United States. doi:10.1145/3077257.3077260.
Tamagnini, Paolo, Krause, Josua W., Dasgupta, Aritra, and Bertini, Enrico. Sun . "Interpreting Black-Box Classifiers Using Instance-Level Visual Explanations". United States. doi:10.1145/3077257.3077260.
@article{osti_1358512,
title = {Interpreting Black-Box Classifiers Using Instance-Level Visual Explanations},
author = {Tamagnini, Paolo and Krause, Josua W. and Dasgupta, Aritra and Bertini, Enrico},
abstractNote = {To realize the full potential of machine learning in diverse real- world domains, it is necessary for model predictions to be readily interpretable and actionable for the human in the loop. Analysts, who are the users but not the developers of machine learning models, often do not trust a model because of the lack of transparency in associating predictions with the underlying data space. To address this problem, we propose Rivelo, a visual analytic interface that enables analysts to understand the causes behind predictions of binary classifiers by interactively exploring a set of instance-level explanations. These explanations are model-agnostic, treating a model as a black box, and they help analysts in interactively probing the high-dimensional binary data space for detecting features relevant to predictions. We demonstrate the utility of the interface with a case study analyzing a random forest model on the sentiment of Yelp reviews about doctors.},
doi = {10.1145/3077257.3077260},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Sun May 14 00:00:00 EDT 2017},
month = {Sun May 14 00:00:00 EDT 2017}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • Finite elements methods have long made use of model order reduction (MOR), particularly in the context of fast freqeucny sweeps. In this paper, we discuss a black-box MOR technique, applicable to a many solution methods and not restricted only to spectral responses. We also discuss automated methods for generating a reduced order model that meets a given error tolerance. Numerical examples demonstrate the effectiveness and wide applicability of the method. With the advent of improved computing hardware and numerous fast solution techniques, the field of computational electromagnetics are progressed rapidly in terms of the size and complexity of problems thatmore » can be solved. Numerous applications, however, require the solution of a problem for many different configurations, including optimization, parameter exploration, and uncertainly quantification, where the parameters that may be changed include frequency, material properties, geometric dimensions, etc. In such cases, thousands of solutions may be needed, so solve times of even a few minutes can be burdensome. Model order reduction (MOR) may alleviate this difficulty by creating a small model that can be evaluated quickly. Many MOR techniques have been applied to electromagnetic problems over the past few decades, particularly in the context of fast frequency sweeps. Recent works have extended these methods to allow more than one parameter and to allow the parameters to represent material and geometric properties. There are still limitations with these methods, however. First, they almost always assume that the finite element method is used to solve the problem, so that the system matrix is a known function of the parameters. Second, although some authors have presented adaptive methods (e.g., [2]), the order of the model is often determined before the MOR process begins, with little insight about what order is actually needed to reach the desired accuracy. Finally, it not clear how to efficiently extend most methods to the multiparameter case. This paper address the above shortcomings be developing a method that uses a block-box approach to the solution method, is adaptive, and is easily extensible to many parameters.« less
  • Recent work in supervised learning has shown that a surprisingly simple Bayesian classifier with strong assumptions of independence among features, called naive Bayes, is competitive with state of the art classifiers such as C4.5. This fact raises the question of whether a classifier with less restrictive assumptions can perform even better. In this paper we examine and evaluate approaches for inducing classifiers from data, based on recent results in the theory of learning Bayesian networks. Bayesian networks are factored representations of probability distributions that generalize the naive Bayes classifier and explicitly represent statements about independence. Among these approaches we singlemore » out a method we call Tree Augmented Naive Bayes (TAN), which outperforms naive Bayes, yet at the same time maintains the computational simplicity (no search involved) and robustness which are characteristic of naive Bayes. We experimentally tested these approaches using benchmark problems from the U. C. Irvine repository, and compared them against C4.5, naive Bayes, and wrapper-based feature selection methods.« less
  • Abstract not provided.