TIP: Typifying the Interpretability of Procedures

Abstract

We provide a novel notion of what it means to be interpretable, looking past the usual association with human understanding. Our key insight is that interpretability is not an absolute concept and so we define it relative to a target model, which may or may not be a human. We define a framework that allows for comparing interpretable procedures by linking them to important practical aspects such as accuracy and robustness. We characterize many of the current state-of-the-art interpretable methods in our framework portraying its general applicability. Finally, principled interpretable strategies are proposed and empirically evaluated on synthetic data, as well as on the largest public olfaction dataset that was made recently available \cite{olfs}. We also experiment on MNIST with a simple target model and different oracle models of varying complexity. This leads to the insight that the improvement in the target model is not only a function of the oracle model's performance, but also its relative complexity with respect to the target model. Further experiments on CIFAR-10, a real manufacturing dataset and FICO dataset showcase the benefit of our methods over Knowledge Distillation when the target models are simple and the complex model is a neural network.

References in corpus (8)

• Distilling the Knowledge in a Neural Network
• Towards A Rigorous Science of Interpretable Machine Learning
• Methods for Interpreting and Understanding Deep Neural Networks
• Synthesizing the preferred inputs for neurons in neural networks via deep generator networks
• Multifaceted Feature Visualization: Uncovering the Different Types of Features Learned By Each Neuron in Deep Neural Networks
• Interpreting Blackbox Models via Model Extraction
• Interpretable Two-level Boolean Rule Learning for Classification
• Learning with Changing Features

Cited by in corpus (10)

• Interpretable Machine Learning -- A Brief History, State-of-the-Art and Challenges
• Explainable Artificial Intelligence Approaches: A Survey
• Interpretable to Whom? A Role-based Model for Analyzing Interpretable Machine Learning Systems
• Quantifying Model Complexity via Functional Decomposition for Better Post-Hoc Interpretability
• Towards Quantification of Explainability in Explainable Artificial Intelligence Methods
• Improving Simple Models with Confidence Profiles
• Latent-CF: A Simple Baseline for Reverse Counterfactual Explanations
• Nonlinear Semi-Parametric Models for Survival Analysis
• A Tour of Convolutional Networks Guided by Linear Interpreters
• Explanation from Specification