Causal Interpretability for Machine Learning - Problems, Methods and Evaluation (2020-03-09T00:00:00.000000Z)

TL;DR

This work presents a comprehensive survey on causal interpretable models from the aspects of the problems and methods and provides in-depth insights into the existing evaluation metrics for measuring interpretability, which can help practitioners understand for what scenarios each evaluation metric is suitable.

Abstract

Machine learning models have had discernible achievements in a myriad of applications. However, most of these models are black-boxes, and it is obscure how the decisions are made by them. This makes the models unreliable and untrustworthy. To provide insights into the decision making processes of these models, a variety of traditional interpretable models have been proposed. Moreover, to generate more humanfriendly explanations, recent work on interpretability tries to answer questions related to causality such as "Why does this model makes such decisions?" or "Was it a specific feature that caused the decision made by the model?". In this work, models that aim to answer causal questions are referred to as causal interpretable models. The existing surveys have covered concepts and methodologies of traditional interpretability. In this work, we present a comprehensive survey on causal interpretable models from the aspects of the problems and methods. In addition, this survey provides in-depth insights into the existing evaluation metrics for measuring interpretability, which can help practitioners understand for what scenarios each evaluation metric is suitable.

Authors

Raha Moraffah

1 papers

Mansooreh Karami

1 papers

Ruocheng Guo

2 papers

TL;DR

Abstract

Authors

References112 items

Feature Selection

Causal Learning in Question Quality Improvement

Interpretable Machine Learning

PC-Fairness: A Unified Framework for Measuring Causality-based Fairness

Explaining Visual Models by Causal Attribution

A Survey on Bias and Fairness in Machine Learning

Explaining Classifiers with Causal Concept Effect (CaCE)

Evaluating Explanation Without Ground Truth in Interpretable Machine Learning

Generative Counterfactual Introspection for Explainable Deep Learning

Causal Interpretations of Black-Box Models

Interpretable Counterfactual Explanations Guided by Prototypes

Explaining Deep Learning Models with Constrained Adversarial Examples

Generating Counterfactual and Contrastive Explanations using SHAP

Explainable Reinforcement Learning Through a Causal Lens

Explaining machine learning classifiers through diverse counterfactual explanations

On the Accuracy of Influence Functions for Measuring Group Effects

Introduction to Data Mining

Counterfactual Visual Explanations

The seven tools of causal inference, with reflections on machine learning

Neural Network Attributions: A Causal Perspective

Learning Interpretable Models with Causal Guarantees

Counterfactuals uncover the modular structure of deep generative models

Multimodal Explanations by Predicting Counterfactuality in Videos

Incorporating Interpretability into Latent Factor Models via Fast Influence Analysis

Interpretable Credit Application Predictions With Counterfactual Explanations

Explaining Deep Learning Models using Causal Inference

Sanity Checks for Saliency Maps

GAN Dissection: Visualizing and Understanding Generative Adversarial Networks

A Survey of Learning Causality with Data

Fairness through Causal Awareness: Learning Causal Latent-Variable Models for Biased Data

Techniques for interpretable machine learning

Model Agnostic Supervised Local Explanations

Generating Counterfactual Explanations with Natural Language

Explaining Explanations: An Overview of Interpretability of Machine Learning

Fairness in Decision-Making - The Causal Explanation Formula

The Building Blocks of Interpretability

Causal Learning and Explanation of Deep Neural Networks via Autoencoded Activations

Interpreting CNNs via Decision Trees

Theoretical Impediments to Machine Learning With Seven Sparks from the Causal Revolution

Visual interpretability for deep learning: a survey

Distilling a Neural Network Into a Soft Decision Tree

Graph Attention Networks

Interpretation of Neural Networks is Fragile

Counterfactual Explanations Without Opening the Black Box: Automated Decisions and the GDPR

A causal framework for explaining the predictions of black-box sequence-to-sequence models

Efficient Data Representation by Selecting Prototypes with Importance Weights

Explanation in Artificial Intelligence: Insights from the Social Sciences

Attention is All you Need

Avoiding Discrimination through Causal Reasoning

A Unified Approach to Interpreting Model Predictions

Counterfactual Fairness

Understanding Black-box Predictions via Influence Functions

Towards A Rigorous Science of Interpretable Machine Learning

Knowing When to Look: Adaptive Attention via a Visual Sentinel for Image Captioning

beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework

Fair Prediction with Disparate Impact: A Study of Bias in Recidivism Prediction Instruments

Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization

False Positives, False Negatives, and False Analyses: A Rejoinder to "Machine Bias: There's Software Used across the Country to Predict Future Criminals. and It's Biased against Blacks"

European Union Regulations on Algorithmic Decision-Making and a "Right to Explanation"

EU regulations on algorithmic decision-making and a "right to explanation"

Model-Agnostic Interpretability of Machine Learning

Estimating individual treatment effect: generalization bounds and algorithms

Hierarchical Attention Networks for Document Classification

InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets

The mythos of model interpretability

Hierarchical Question-Image Co-Attention for Visual Question Answering

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

The Limitations of Deep Learning in Adversarial Settings

Ask, Attend and Answer: Exploring Question-Guided Spatial Attention for Visual Question Answering

DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks

Stacked Attention Networks for Image Question Answering

Intelligible Models for HealthCare: Predicting Pneumonia Risk and Hospital 30-day Readmission

Distilling the Knowledge in a Neural Network

Show, Attend and Tell: Neural Image Caption Generation with Visual Attention

Striving for Simplicity: The All Convolutional Net

Explaining and Harnessing Adversarial Examples