Neural Additive Models: Interpretable Machine Learning with Neural Nets

Abstract

Deep neural networks (DNNs) are powerful black-box predictors that have achieved impressive performance on a wide variety of tasks. However, their accuracy comes at the cost of intelligibility: it is usually unclear how they make their decisions. This hinders their applicability to high stakes decision-making domains such as healthcare. We propose Neural Additive Models (NAMs) which combine some of the expressivity of DNNs with the inherent intelligibility of generalized additive models. NAMs learn a linear combination of neural networks that each attend to a single input feature. These networks are trained jointly and can learn arbitrarily complex relationships between their input feature and the output. Our experiments on regression and classification datasets show that NAMs are more accurate than widely used intelligible models such as logistic regression and shallow decision trees. They perform similarly to existing state-of-the-art generalized additive models in accuracy, but are more flexible because they are based on neural nets instead of boosted trees. To demonstrate this, we show how NAMs can be used for multitask learning on synthetic data and on the COMPAS recidivism data due to their composability, and demonstrate that the differentiability of NAMs allows them to train more complex interpretable models for COVID-19.

Spotlight (Top 3%) at NeurIPS 2021

References in corpus (2)

• Practical Bayesian Optimization of Machine Learning Algorithms
• A Closer Look at Memorization in Deep Networks

Cited by in corpus (19)

• Opportunities and Challenges in Explainable Artificial Intelligence (XAI): A Survey
• Logic Explained Networks
• Augmenting Interpretable Models with LLMs during Training
• Unwrapping The Black Box of Deep ReLU Networks: Interpretability, Diagnostics, and Simplification
• Coalesced Multi-Output Tsetlin Machines with Clause Sharing
• GAMI-Net: An Explainable Neural Network based on Generalized Additive Models with Structured Interactions
• Interpretable Machine Learning: Moving From Mythos to Diagnostics
• Combining Graph Neural Networks and Spatio-temporal Disease Models to Predict COVID-19 Cases in Germany
• Link Prediction using Graph Neural Networks for Master Data Management
• Mixture of Linear Models Co-supervised by Deep Neural Networks
• Neural Mixture Distributional Regression
• Robust Semantic Interpretability: Revisiting Concept Activation Vectors
• Explainable Recommendation Systems by Generalized Additive Models with Manifest and Latent Interactions
• Partially Interpretable Estimators (PIE): Black-Box-Refined Interpretable Machine Learning
• An Upper Limit of Decaying Rate with Respect to Frequency in Deep Neural Network
• NOTMAD: Estimating Bayesian Networks with Sample-Specific Structures and Parameters
• It's FLAN time! Summing feature-wise latent representations for interpretability
• RECA-PD: A Robust Explainable Cross-Attention Method for Speech-based Parkinson's Disease Classification
• Neural Additive and Basis Models with Feature Selection and Interactions