mlsanityData to 7 Oct 2026

Feature Selection with Distance Correlation

arXiv 2212.00046 ↗First posted 30 Nov 2022hep-ph cs.LG hep-excited by 3

Not on the map. Why?

Abstract (arXiv)

Choosing which properties of the data to use as input to multivariate decision algorithms -- a.k.a. feature selection -- is an important step in solving any problem with machine learning. While there is a clear trend towards training sophisticated deep networks on large numbers of relatively unprocessed inputs (so-called automated feature engineering), for many tasks in physics, sets of theoretically well-motivated and well-understood features already exist. Working with such features can bring many benefits, including greater interpretability, reduced training and run time, and enhanced stability and robustness. We develop a new feature selection method based on Distance Correlation (DisCo), and demonstrate its effectiveness on the tasks of boosted top- and W-tagging. Using our method to select features from a set of over 7,000 energy flow polynomials, we show that we can match the performance of much deeper architectures, by using only ten features and two orders-of-magnitude fewer model parameters.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 23 of its 71 references that are in the map. First posted on arXiv 30 Nov 2022.

Builds on

23 of 71 references.

References · 23

  1. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  2. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  3. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
  4. 2019Explaining individual predictions when features are dependent: More accurate approximations to Shapley valuescited by 205
  5. 2018Fast inference of deep neural networks in FPGAs for particle physicscited by 132
  6. 2020Lorentz Group Equivariant Neural Network for Particle Physicscited by 120
  7. 2018Energy Flow Networks: Deep Sets for Particle Jetscited by 83
  8. 2019Explainable AI for Trees: From Local Explanations to Global Understandingcited by 64
  9. 2021Machine Learning in the Search for New Fundamental Physicscited by 50
  10. 2015Jet-Images -- Deep Learning Editioncited by 41
  11. 2022PELICAN: Permutation Equivariant and Lorentz Invariant or Covariant Aggregator Network for Particle Physicscited by 26
  12. 2021Point Cloud Transformers applied to Collider Physicscited by 23
11 more
  1. 2017QCD-Aware Recursive Neural Networks for Jet Physicscited by 22
  2. 2020Jet tagging in the Lund plane with graph networkscited by 22
  3. 2013Partial Distance Correlation with Methods for Dissimilaritiescited by 20
  4. 2017Jet Constituents for Deep Neural Network Based Top Quark Taggingcited by 15
  5. 2021Online-compatible Unsupervised Non-resonant Anomaly Detectioncited by 15
  6. 2020Explainable AI for ML jet taggers using expert variables and layerwise relevance propagationcited by 14
  7. 2017Long Short-Term Memory (LSTM) networks with jet constituents for boosted top tagging at the LHCcited by 13
  8. 2019Interpretable Deep Learning for Two-Prong Jet Classification with Jet Spectracited by 7
  9. 2022BIP: Boost Invariant Polynomials for Efficient Jet Taggingcited by 7
  10. 2022A Detailed Study of Interpretability of Deep Neural Network based Top Taggerscited by 6
  11. 2018Infrared Safety of a Neural-Net Top Tagging Algorithmcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 0; 2025: 1; 2026: 1.

3 papers.

  1. 2025A Step Toward Interpretability: Smearing the Likelihoodcited by 1
  2. 202319 Parameters Is All You Need: Tiny Neural Networks for Particle Physicscited by 0
  3. 2026K-IPO: Kendall-constrained Importance Preserving Oversampling for Imbalanced Tabular Datacited by 0