False Negative Rate

False negative rate (FNR) refers to the proportion of actual positives incorrectly identified as negatives, a critical metric across diverse fields demanding high-accuracy classification. Current research focuses on improving FNR estimation and mitigation, particularly in applications with asymmetric costs of false positives and negatives, such as fraud detection and anomaly detection in time-series data. This involves developing novel algorithms, including variational autoencoders and methods leveraging soft labels, to refine model training and reduce bias in negative sampling strategies. Accurate FNR estimation is crucial for optimizing model performance and ensuring reliable decision-making in various applications, from e-commerce to wildfire detection and robotics.

Papers

July 9, 2024

TeVAE: A Variational Autoencoder Approach for Discrete Online Anomaly Detection in Variable-state Multivariate Time-series Data
Lucas Correia, Jan-Christoph Goos, Philipp Klein, Thomas Bäck, Anna V. Kononova
Anomaly Detection Variational Autoencoder Unlabeled Anomaly Multivariate Time Series Data False Negative Rate

January 27, 2024

Data-Driven Estimation of the False Positive Rate of the Bayes Binary Classifier via Soft Labels
Minoh Jeong, Martina Cardone, Alex Dytso
Binary Classification Soft Label Optimal Classifier Bayes Classifier False Positive Rate False Negative Rate

November 11, 2023

Mitigating Pooling Bias in E-commerce Search via False Negative Estimation
Xiaochen Wang, Xiao Xiao, Ruhan Zhang, Xuan Zhang, Taesik Na, Tejaswi Tenneti, Haixun Wang, Fenglong Ma
Negative Sampling E Commerce Search Relevance Map False Negative Rate

August 4, 2023

FPR Estimation for Fraud Detection in the Presence of Class-Conditional Label Noise
Justin Tittelfitz
Label Noise Speech Presence Fraud Detection Binary Classifier Class Conditional False Negative Rate

November 17, 2022

Online Distribution Shift Detection via Recency Prediction
Rachel Luo, Rohan Sinha, Yixiao Sun, Ali Hindy, Shengjia Zhao, Silvio Savarese, Edward Schmerling, Marco Pavone
Distribution Shift Robot Learning Efficient Detection Rapid Detection False Negative Rate Recency Heuristic Distribution Shift Detection

November 3, 2022

Image-based Early Detection System for Wildfires
Omkar Ranadive, Jisu Kim, Serin Lee, Youngseo Cha, Heechan Park, Minkook Cho, Young K. Hwang
Automatic Detection Improved Detection Wildfire Detection Forest Fire False Negative Rate Wildfire Smoke

May 11, 2022

Is calibration a fairness requirement? An argument from the point of view of moral philosophy and decision theory
Michele Loi, Christoph Heitz
Calibration Performance Group Fairness Algorithmic Fairness Argument Quality Decision Theory Moral Philosophy Statistical Fairness False Negative Rate Fairness Condition

April 13, 2022

Stealing and Evading Malware Classifiers and Antivirus at Low False Positive Conditions
Maria Rigaki, Sebastian Garcia
Surrogate Model Malware Detection Malware Attack Adversarial Malware Malware Classifier False Negative Rate

February 15, 2022

Conformal Prediction Sets with Limited False Positives
Adam Fisch, Tal Schuster, Tommi Jaakkola, Regina Barzilay
Human Prediction Conformal Prediction Conformal Prediction Set Credal Set False Negative Rate

False Negative Rate

Papers

TeVAE: A Variational Autoencoder Approach for Discrete Online Anomaly Detection in Variable-state Multivariate Time-series Data

Data-Driven Estimation of the False Positive Rate of the Bayes Binary Classifier via Soft Labels

Mitigating Pooling Bias in E-commerce Search via False Negative Estimation

FPR Estimation for Fraud Detection in the Presence of Class-Conditional Label Noise

Online Distribution Shift Detection via Recency Prediction

Image-based Early Detection System for Wildfires

Is calibration a fairness requirement? An argument from the point of view of moral philosophy and decision theory

Stealing and Evading Malware Classifiers and Antivirus at Low False Positive Conditions

Conformal Prediction Sets with Limited False Positives