283 citations · 283 across the 2 of their papers we have counts for
Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
A Strong Balanced-Softmax Classifier-Retraining Baseline for Long-Tailed Recognition
Juan Terven, Diana Margarita Córdova Esparza, Julio Alejandro Romero Gonzalez +4
Long-tailed recognition methods often modify losses, margins, or representations to reduce the dominance of frequent classes. We ask whether, after Balanced Softmax training, the r…
cs.LG2023★ 283 cited
Loss Functions and Metrics in Deep Learning
Juan Terven, Diana M. Cordova-Esparza, Alfonso Ramirez-Pedraza +2
This paper presents a comprehensive review of loss functions and performance metrics in deep learning, highlighting key developments and practical insights across diverse applicati…