Skip to content
DraconiticResearch
Search Sources API
201,683 studies31 sourcesUpdated 42 min ago
PreprintCC BY 4.0arXiv

Extreme Binary Classification: Extreme Value Theory for Extreme Constraint on False Negative

  • Samuel Gruffaz
  • Muhammad Fawad
  • Jaakko Nevalainen
Open at sourcePDFDOICite

Abstract

While binary classification is one of the most extensively studied problems in machine learning, the regime in which the goal is to learn a classifier with an almost zero false negative rate remains largely unexplored. In this paper, we introduce the Extreme Binary Classification problem, where the objective is to learn a classifier whose false negative rate α is constrained by εN1=oN1→∞(1/N1), with N1 denoting the number of positive examples in the training set. To address this problem, we propose a threshold adaptation method theoretically grounded in guarantees derived from Extreme Value Theory, together with a feature selection procedure based on a permutation test applied to sample maxima. Experimental results on four real-world datasets of varying sizes demonstrate that our approach compares favorably with state-of-the-art methods. In addition, we illustrate its interpretability through an application to a cancer screening dataset.

Related

  1. From Gradient Descent to Potential Theory: A Geometric Dictionary for Binary Classification
    Catalin Vasii
    arXivPreprintCC BY 4.0Full textGeometry & topologyarXiv
  2. Polynomially Solvable Cases of the Quadratic Binary Optimization Problem with a Fixed Cardinality Constraint
    Thi Thanh Tu Le, Thorsten Koch
    arXivPreprintCC BY-NC-SA 4.0Full textOptimisation & controlarXiv
  3. Ground States of a Coupled Fractional Schrödinger System under a Total-Mass Constraint: Complete Classification and Concentration Behavior
    Lijun Liu, Jiabin Zuo
    arXivPreprintCC0Full textAnalysis & PDEsarXiv
  4. A complete classification of the existence of finite-dimensional quantum solutions to inconsistent linear constraint systems
    Markus Frembs
    arXivPreprintCC BY 4.0Full textMathematical physicsQuantum physics & quantum informationarXiv
  5. Classification of binary, ternary and quaternary Hermitian LCD codes
    Keita Ishizuka
    arXivPreprintCC BY 4.0Full textGeneral mathematicsarXiv

Cite

BibTeXRIS
BibTeX
@misc{gruffaz2026extreme,
  title = {{Extreme Binary Classification: Extreme Value Theory for Extreme Constraint on False Negative}},
  author = {Samuel Gruffaz and Muhammad Fawad and Jaakko Nevalainen},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.09984},
  eprint = {2610.09984},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.09984},
  note = {CC BY 4.0},
}
RIS
TY  - UNPB
TI  - Extreme Binary Classification: Extreme Value Theory for Extreme Constraint on False Negative
AU  - Samuel Gruffaz
AU  - Muhammad Fawad
AU  - Jaakko Nevalainen
PY  - 2026
DA  - 2026/10/07
PB  - arXiv
DO  - 10.48550/arxiv.2610.09984
UR  - https://arxiv.org/abs/2610.09984
AB  - While binary classification is one of the most extensively studied problems in machine learning, the regime in which the goal is to learn a classifier with an almost zero false negative rate remains largely unexplored. In this paper, we introduce the Extreme Binary Classification problem, where the objective is to learn a classifier whose false negative rate α is constrained by ε_N_1=o_N_1→∞(1/N_1), with N_1 denoting the number of positive examples in the training set. To address this problem, we propose a threshold adaptation method theoretically grounded in guarantees derived from Extreme Value Theory, together with a feature selection procedure based on a permutation test applied to sample maxima. Experimental results on four real-world datasets of varying sizes demonstrate that our approach compares favorably with state-of-the-art methods. In addition, we illustrate its interpretability through an application to a cancer screening dataset.
ER  -