Classification without labels: Learning from mixed samples in high energy physics

2017-08-09 · Eric M. Metodiev, Benjamin Nachman, Jesse Thaler

Modern machine learning techniques can be used to construct powerful models for difficult collider physics problems. In many applications, however, these models are trained on imperfect simulations due to a lack of truth-level information in the data, which risks the model learning artifacts of the simulation. In this paper, we introduce the paradigm of classification without labels (CWoLa) in which a classifier is trained to distinguish statistical mixtures of classes, which are common in collider physics. Crucially, neither individual labels nor class proportions are required, yet we prove that the optimal classifier in the CWoLa paradigm is also the optimal classifier in the traditional fully-supervised case where all label information is available. After demonstrating the power of this method in an analytical toy example, we consider a realistic benchmark for collider physics: distinguishing quark- versus gluon-initiated jets using mixed quark/gluon training samples. More generally, CWoLa can be applied to any classification problem where labels or class proportions are unknown or simulations are unreliable, but statistical mixtures of the classes are available.

📄 PDF Abstract BibTeX arXiv:1708.02949

Code (1)

hep-lbdl/gaiacwola tf

Tasks

General Classification Vocal Bursts Intensity Prediction

BnSentMix: A Diverse Bengali-English Code-Mixed Dataset for Sentiment Analysis

2024-08-16 · Sadia Alam, Md Farhan Ishmam, Navid Hasin Alvee, Md Shahnewaz Siddique 외

The widespread availability of code-mixed data can provide valuable insights into low-resource languages like Bengali, which have limited datasets. Sentiment analysis has been a fundamental text classification task acros…

Diversity Sentiment Analysis Sentiment Classification text-classification+1

AutoMix: Unveiling the Power of Mixup for Stronger Classifiers

2021-03-24 · Zicheng Liu, Siyuan Li, Di wu, Zihan Liu 외

Data mixing augmentation have proved to be effective in improving the generalization ability of deep neural networks. While early methods mix samples by hand-crafted policies (e.g., linear interpolation), recent methods …

Classification Data Augmentation General Classification Image Classification+1

Weakly-Supervised Anomaly Detection in the Milky Way

2023-05-05 · Mariel Pettee, Sowmya Thanvantri, Benjamin Nachman, David Shih 외

Large-scale astrophysics datasets present an opportunity for new machine learning techniques to identify regions of interest that might otherwise be overlooked by traditional searches. To this end, we use Classification …

Anomaly Detection Supervised Anomaly Detection Weakly-supervised Anomaly Detection

Harnessing Hard Mixed Samples with Decoupled Regularizer

2022-03-21 · NeurIPS 2023 11 · Zicheng Liu, Siyuan Li, Ge Wang, Cheng Tan 외

Mixup is an efficient data augmentation approach that improves the generalization of neural networks by smoothing the decision boundary with mixed data. Recently, dynamic mixup methods have improved previous static polic…

Data Augmentation

SelectMix: Enhancing Label Noise Robustness through Targeted Sample Mixing

2025-09-14 · Qiuhao Liu, Ling Li, Yao Lu, Qi Xuan 외 arxiv

Deep neural networks tend to memorize noisy labels, severely degrading their generalization performance. Although Mixup has demonstrated effectiveness in improving generalization and robustness, existing Mixup-based meth…

Learning with noisy labels