Despite widespread adoption, machine learning models remain mostly black boxes. Understanding the reasons behind predictions is, however, quite important in assessing trust, which is fundamental if one plans to take action based on a prediction, or when choosing whether to deploy a new model. Such understanding also provides insights into the model, which can be used to transform an untrustworthy model or prediction into a trustworthy one. In this work, we propose LIME, a novel explanation technique that explains the predictions of any classifier in an interpretable and faithful manner, by learning an interpretable model locally varound the prediction. We also propose a method to explain models by presenting representative individual predictions and their explanations in a non-redundant way, framing the task as a submodular optimization problem. We demonstrate the flexibility of these methods by explaining different models for text (e.g. random forests) and image classification (e.g. neural networks). We show the utility of explanations via novel experiments, both simulated and with human subjects, on various scenarios that require trust: deciding if one should trust a prediction, choosing between models, improving an untrustworthy classifier, and identifying why a classifier should not be trusted.

/pdf/why-should-i-trust-you-explaining-the-predictions-of-any-ejtlrtykxr.pdf

"Why Should I Trust You?": Explaining the Predictions of Any Classifier

How can we explain the predictions of a black-box model? In this paper, we use influence functions -- a classic technique from robust statistics -- to trace a model's prediction through the learning algorithm and back to its training data, thereby identifying training points most responsible for a given prediction. To scale up influence functions to modern machine learning settings, we develop a simple, efficient implementation that requires only oracle access to gradients and Hessian-vector products. We show that even on non-convex and non-differentiable models where the theory breaks down, approximations to influence functions can still provide valuable information. On linear models and convolutional neural networks, we demonstrate that influence functions are useful for multiple purposes: understanding model behavior, debugging models, detecting dataset errors, and even creating visually-indistinguishable training-set attacks.

Understanding Black-box Predictions via Influence Functions

我的台灣, 看見心靈的故鄉 =2009林磐聳藝術與設計展

In diesem Beitrag verdeutliche ich meine Besorgnis uber zahlreiche Umformungen der Grounded Theory (GT), die mit deren Rezeption durch Methodologien der qualitative Datenanalyse (QDA) einhergehen, und hieraus folgende Erosionen. Ich skizziere zunachst einige Beispiele hierfur, um danach die essentiellen Bestandteile der klassischen GT-Methodologie zusammenzufassen. Ich hoffe, dass dieser Beitrag meine Besorgnis uber die wachsende, aber meines Erachtens missverstandene Einvernahme von GT durch QDA-Methodologien veranschaulicht und zugleich als einfuhrender Leitfaden fur Novizen und Novizinnen dient, die daran interessiert sind, die grundlegenden Prinzipien der GT nachzuvollziehen. \nURN: urn:nbn:de:0114-fqs040245

Remodeling Grounded Theory

How can we explain the predictions of a black-box model? In this paper, we use influence functions — a classic technique from robust statistics — to trace a model's prediction through the learning algorithm and back to its training data, thereby identifying training points most responsible for a given prediction. To scale up influence functions to modern machine learning settings, we develop a simple, efficient implementation that requires only oracle access to gradients and Hessian-vector products. We show that even on non-convex and non-differentiable models where the theory breaks down, approximations to influence functions can still provide valuable information. On linear models and convolutional neural networks, we demonstrate that influence functions are useful for multiple purposes: understanding model behavior, debugging models, detecting dataset errors, and even creating visually-indistinguishable training-set attacks.

/pdf/understanding-black-box-predictions-via-influence-functions-285r2tv1ec.pdf

Understanding black-box predictions via influence functions

Advances in artificial intelligence (AI) frame opportunities and challenges for user interface design. Principles for human-AI interaction have been discussed in the human-computer interaction community for over two decades, but more study and innovation are needed in light of advances in AI and the growing uses of AI technologies in human-facing applications. We propose 18 generally applicable design guidelines for human-AI interaction. These guidelines are validated through multiple rounds of evaluation including a user study with 49 design practitioners who tested the guidelines against 20 popular AI-infused products. The results verify the relevance of the guidelines over a spectrum of interaction scenarios and reveal gaps in our knowledge, highlighting opportunities for further research. Based on the evaluations, we believe the set of design guidelines can serve as a resource to practitioners working on the design of applications and features that harness AI technologies, and to researchers interested in the further development of human-AI interaction design principles.

Guidelines for Human-AI Interaction

We demonstrate the value of collecting semantic parse labels for knowledge base question answering. In particular, (1) unlike previous studies on small-scale datasets, we show that learning from labeled semantic parses significantly improves overall performance, resulting in absolute 5 point gain compared to learning from answers, (2) we show that with an appropriate user interface, one can obtain semantic parses with high accuracy and at a cost comparable or lower than obtaining just answers, and (3) we have created and shared the largest semantic-parse labeled dataset to date in order to advance research in question answering.

/pdf/the-value-of-semantic-parse-labeling-for-knowledge-base-55x7zuh2yy.pdf

The Value of Semantic Parse Labeling for Knowledge Base Question Answering

Model building in machine learning is an iterative process. The performance analysis and debugging step typically involves a disruptive cognitive switch from model building to error analysis, discouraging an informed approach to model building. We present ModelTracker, an interactive visualization that subsumes information contained in numerous traditional summary statistics and graphs while displaying example-level performance and enabling direct error examination and debugging. Usage analysis from machine learning practitioners building real models with ModelTracker over six months shows ModelTracker is used often and throughout model building. A controlled experiment focusing on ModelTracker's debugging capabilities shows participants prefer ModelTracker over traditional tools without a loss in model performance.

/pdf/modeltracker-redesigning-performance-analysis-tools-for-2e3atde9qb.pdf

ModelTracker: Redesigning Performance Analysis Tools for Machine Learning

Performance analysis is critical in applied machine learning because it influences the models practitioners produce. Current performance analysis tools suffer from issues including obscuring important characteristics of model behavior and dissociating performance from data. In this work, we present Squares, a performance visualization for multiclass classification problems. Squares supports estimating common performance metrics while displaying instance-level distribution information necessary for helping practitioners prioritize efforts and access data. Our controlled study shows that practitioners can assess performance significantly faster and more accurately with Squares than a confusion matrix, a common performance analysis tool in machine learning.

/pdf/squares-supporting-interactive-performance-analysis-for-2jsi6cg7tz.pdf

Squares: Supporting Interactive Performance Analysis for Multiclass Classifiers

The current processes for building machine learning systems require practitioners with deep knowledge of machine learning. This significantly limits the number of machine learning systems that can be created and has led to a mismatch between the demand for machine learning systems and the ability for organizations to build them. We believe that in order to meet this growing demand for machine learning systems we must significantly increase the number of individuals that can teach machines. We postulate that we can achieve this goal by making the process of teaching machines easy, fast and above all, universally accessible. 
While machine learning focuses on creating new algorithms and improving the accuracy of "learners", the machine teaching discipline focuses on the efficacy of the "teachers". Machine teaching as a discipline is a paradigm shift that follows and extends principles of software engineering and programming languages. We put a strong emphasis on the teacher and the teacher's interaction with data, as well as crucial components such as techniques and design principles of interaction and visualization. 
In this paper, we present our position regarding the discipline of machine teaching and articulate fundamental machine teaching principles. We also describe how, by decoupling knowledge about machine learning algorithms from the process of teaching, we can accelerate innovation and empower millions of new uses for machine learning models.

/pdf/machine-teaching-a-new-paradigm-for-building-machine-26s8dmz0mw.pdf

Jina Suh

Papers

Guidelines for Human-AI Interaction

The Value of Semantic Parse Labeling for Knowledge Base Question Answering

ModelTracker: Redesigning Performance Analysis Tools for Machine Learning

Squares: Supporting Interactive Performance Analysis for Multiclass Classifiers

Machine Teaching: A New Paradigm for Building Machine Learning Systems.