mlsanityData to 14 Oct 2026

Scalable Interpretability via Polynomials

arXiv 2205.14108 ↗First posted 27 May 2022cs.LG cs.CVcited by 25

Not on the map. Why?

Abstract (arXiv)

Generalized Additive Models (GAMs) have quickly become the leading choice for inherently-interpretable machine learning. However, unlike uninterpretable methods such as DNNs, they lack expressive power and easy scalability, and are hence not a feasible alternative for real-world tasks. We present a new class of GAMs that use tensor rank decompositions of polynomials to learn powerful, em inherently-interpretable models. Our approach, titled Scalable Polynomial Additive Models (SPAM) is effortlessly scalable and models em all higher-order feature interactions without a combinatorial parameter explosion. SPAM outperforms all current interpretable approaches, and matches DNN/XGBoost performance on a series of real-world benchmarks with up to hundreds of thousands of features. We demonstrate by human subject evaluations that SPAMs are demonstrably more interpretable in practice, and are hence an effortless replacement for DNNs for creating interpretable and high-performance systems suitable for large-scale machine learning. Source code is available at https://github.com/facebookresearch/nbm-spam.

Key facts. Cited by 25 papers in the map (data through 14 Oct 2026). Builds on 23 of its 81 references that are in the map. First posted on arXiv 27 May 2022.

Builds on

23 of 81 references.

References · 23

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  3. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  4. 2017Learning Important Features Through Propagating Activation Differencescited by 1,458
  5. 2020Concept Bottleneck Modelscited by 1,032
  6. 2018This Looks Like That: Deep Learning for Interpretable Image Recognitioncited by 737
  7. 2019TabNet: Attentive Interpretable Tabular Learningcited by 648
  8. 2017Interpretation of Neural Networks is Fragilecited by 444
  9. 2020Neural Additive Models: Interpretable Machine Learning with Neural Netscited by 319
  10. 2016Not Just a Black Box: Learning Important Features Through Propagating Activation Differencescited by 297
  11. 2018Metrics for Explainable AI: Challenges and Prospectscited by 260
  12. 2019InterpretML: A Unified Framework for Machine Learning Interpretabilitycited by 246
11 more
  1. 2018Techniques for Interpretable Machine Learningcited by 238
  2. 2021Benchmarking Representation Learning for Natural World Image Collectionscited by 168
  3. 2021Post-hoc Interpretability for Neural NLP: A Surveycited by 156
  4. 2017ParlAI: A Dialog Research Software Platformcited by 139
  5. 2020Reliable Post hoc Explanations: Modeling Uncertainty in Explainabilitycited by 125
  6. 2020Invertible Concept-based Explanations for CNN Models with Non-negative Concept Activation Vectorscited by 120
  7. 2019"How do I fool you?": Manipulating User Trust via Misleading Black Box Explanationscited by 111
  8. 2019The Dangers of Post-hoc Interpretability: Unjustified Counterfactual Explanationscited by 97
  9. 2021NODE-GAM: Neural Generalized Additive Model for Interpretable Deep Learningcited by 88
  10. 2020Pi-nets: Deep Polynomial Neural Networkscited by 39
  11. 2019PolyGAN: High-Order Polynomial Generatorscited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 1; 2023: 8; 2024: 8; 2025: 6; 2026: 2.

25 papers.

Sort:

By area:

  1. 2022"Help Me Help the AI": Understanding How Explainability Can Support Human-AI Interactioncited by 63
  2. 2023Neural Additive Models for Location Scale and Shape: A Framework for Interpretable Neural Regression Beyond the Meancited by 18
  3. 2025A Comprehensive Survey on Self-Interpretable Neural Networkscited by 18
  4. 2025Beyond Linear Probes: Dynamic Safety Monitoring for Language Modelscited by 16
  5. 2023Backpack Language Modelscited by 15
  6. 2023Regularization of polynomial networks for image recognitioncited by 13
  7. 2023Curve Your Enthusiasm: Concurvity Regularization in Differentiable Generalized Additive Modelscited by 12
  8. 2024Multilinear Operator Networkscited by 12
  9. 2023Multi-Objective Optimization of Performance and Interpretability of Tabular Supervised Machine Learning Modelscited by 6
  10. 2023FocusLearn: Fully-Interpretable, High-Performance Modular Neural Networks for Time Seriescited by 5
  11. 2023A Survey of Explainable AI in Deep Visual Modeling: Methods and Metricscited by 4
  12. 2024A Tensor Decomposition Perspective on Second-order RNNscited by 3
13 more
  1. 2025A Training Framework for Optimal and Stable Training of Polynomial Neural Networkscited by 3
  2. 2024Transparent Networks for Multivariate Time Seriescited by 2
  3. 2025Beyond Black-Box Predictions: Identifying Marginal Feature Effects in Tabular Transformer Networkscited by 2
  4. 2026Selecting Feature Interactions for Generalized Additive Models by Distilling Foundation Modelscited by 2
  5. 2023UFO: A unified method for controlling Understandability and Faithfulness Objectives in concept-based explanations for CNNscited by 1
  6. 2024Neural Additive Image Model: Interpretation through Interpolationcited by 1
  7. 2024How Video Meetings Change Your Expressioncited by 1
  8. 2024GAMformer: Bridging Tabular Foundation Models and Interpretable Machine Learningcited by 1
  9. 2024Generalized Sparse Additive Model with Unknown Link Functioncited by 1
  10. 2024Tensor Polynomial Additive Modelcited by 0
  11. 2025LIT-LVM: Structured Regularization for Interaction Terms in Linear Predictors using Latent Variable Modelscited by 0
  12. 2025The Impact of Concept Explanations and Interventions on Human-Machine Collaborationcited by 0
  13. 2026The Quadrilateral Loss: Additivity as a Measurable Behavior of Dense Neural Networkscited by 0