Definitions, methods, and applications in interpretable machine learning
Explore this paper's citation graph
Summary
This work defines interpretability in the context of machine learning and introduces the predictive, descriptive, relevant (PDR) framework for discussing interpretations, and introduces 3 overarching desiderata for evaluation: predictive accuracy, descriptive accuracy, and relevancy.
- Type
- article
- Published
- 2019-01-14
- Cited by
- 1,856
- References
- 112
- Access
- Open access
- OpenAlex
- https://openalex.org/W2910705748
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:204755862
Keywords
Interpretability, Computer science, Artificial intelligence, Categorization, Machine learning
References
- Case-based explanation of non-case-based learning methods
- Probabilistic Graphical Models: Principles and Techniques - Adaptive Computation and Machine Learning
- Understanding Intra-Class Knowledge Inside CNN
- Causal Inference for Statistics, Social, and Biomedical Sciences: An Introduction
- CRITICAL QUESTIONS FOR BIG DATA
- ggplot2: Elegant Graphics for Data Analysis
- Visualizing and Understanding Recurrent Networks
- The Emergence of Machine Learning Techniques in Criminology
- Daytime Arctic Cloud Detection Based on Multi-Angle Satellite Data With Case Studies
- Intelligible Models for HealthCare: Predicting Pneumonia Risk and Hospital 30-day Readmission
- Robust statistics: the approach based on influence functions
- Statistical models and shoe leather
- Relations Between Two Sets of Variates
- The structure and function of explanations.
- Enriched random forests
- PREDICTIVE LEARNING VIA RULE ENSEMBLES
- Classification and regression trees
- IPython: A System for Interactive Scientific Computing
- Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author)
- Toward a Unified Theory of Visual Area V4
Cited by
- Where does Haydn end and Mozart begin? Composer classification of string quartets
- Modeling Stated Preference for Mobility-on-Demand Transit: A Comparison of Machine Learning and Logit Models
- Modeling Heterogeneity in Mode-Switching Behavior Under a Mobility-on-Demand Transit System: An Interpretable Machine Learning Approach
- ExplaiNE: An Approach for Explaining Network Embedding-based Link Predictions
- Model-Agnostic Counterfactual Explanations for Consequential Decisions
- Explainable Machine Learning for Scientific Insights and Discoveries
- Disentangled Attribution Curves for Interpreting Random Forests and Boosted Trees
- LGM-Net: Learning to Generate Matching Networks for Few-Shot Learning
- Hangul Fonts Dataset: a Hierarchical and Compositional Dataset for Interrogating Learned Representations
- Dissecting the binding mechanisms of transcription factors to DNA using a statistical thermodynamics framework
- Understanding artificial intelligence ethics and safety
- Global Aggregations of Local Explanations for Black Box models
- Evaluating Explanation Without Ground Truth in Interpretable Machine Learning
- The mass, fake news, and cognition security
- DeepDrawing: A Deep Learning Approach to Graph Drawing
- Adversarial explanations for understanding image classification decisions and improved neural network robustness
- SIRUS: making random forests interpretable
- Shapley Decomposition of R-Squared in Machine Learning Models
- A Debiased MDI Feature Importance Measure for Random Forests
- Fooling Neural Network Interpretations via Adversarial Model Manipulation
Related papers
- ML Interpretability: Simple Isn't Easy
- When consumers need more interpretability of artificial intelligence (AI) recommendations? The effect of decision-making domains
- Effects of mixed sample data augmentation on interpretability of neural networks
- Assessing the Interpretability of Programmatic Policies with Large Language Models