Weighted Loss

Weighted loss functions are increasingly used to improve the performance of machine learning models by assigning different importance to various training samples or classes. Current research focuses on optimizing weight assignments to address challenges like imbalanced datasets, outlier sensitivity, and interference from data training loops, often employing techniques such as Gaussian weighting, entropy-based scaling, or data-dependent weighting schemes within various model architectures including convolutional neural networks, transformers, and support vector machines. These advancements enhance model robustness, generalization, and efficiency across diverse applications, from time series forecasting and depth estimation to natural language processing and medical image analysis.

Papers

June 19, 2024

Robust Time Series Forecasting with Non-Heavy-Tailed Gaussian Loss-Weighted Sampler
Jiang You, Arben Cela, René Natowicz, Jacob Ouanounou, Patrick Siarry
Multivariate Time Series Forecasting Weighted Loss Robust Forecasting Tailed Loss

April 11, 2024

Depth Estimation using Weighted-loss and Transfer Learning
Muhammad Adeel Hafeez, Michael G. Madden, Ganesh Sistu, Ihsan Ullah
Transfer Learning Depth Estimation Depth Estimation Model New York University Depth V2 Weighted Loss

October 30, 2023

MiLe Loss: a New Loss for Mitigating the Bias of Learning Difficulties in Generative Language Models
Zhenpeng Su, Xing Wu, Xue Bai, Zijia Lin, Hui Chen, Guiguang Ding, Wei Zhou, Songlin Hu
Language Model Absolute Stance Bias Generative Language Model Interest Loss Weighted Loss Easy to Learn Token

October 26, 2023

Tackling Interference Induced by Data Training Loops in A/B Tests: A Weighted Training Approach
Nian Si
Catastrophic Interference Weight Freezing Training Machine Learning Model Training Distribution Weighted Loss B Test Training Loop

July 17, 2023

MuLMINet: Multi-Layer Multi-Input Transformer Network with Weighted Loss
Minwoo Seong, Jeongseok Oh, SeungJun Kim
Sport Analytics Badminton Report Weighted Loss Turn Based

May 6, 2023

A Sea-Land Clutter Classification Framework for Over-the-Horizon-Radar Based on Weighted Loss Semi-supervised GAN
Xiaoxuan Zhang, Zengfu Wang, Kun Lu, Quan Pan, Yang Li
GAN Model Radar Signal Unlabeled Sample Infrared Clutter Background Weighted Loss Land Clutter

February 22, 2023

A Generalized Weighted Loss for SVC and MLP
Filippo Portera
Regression Task Binary Classifier Single Scene Specific MLP Error Estimation Weighted Loss Regression Network

September 28, 2022

Class-Imbalanced Complementary-Label Learning via Weighted Loss
Meng Wei, Yong Zhou, Zhongnian Li, Xinzheng Xu
Complementary Label Weighted Loss Complementary Label Learning Imbalanced Training

September 7, 2022

Difficulty-Net: Learning to Predict Difficulty for Long-Tailed Recognition
Saptarshi Sinha, Hiroki Ohashi
Long Tailed Long Tailed Recognition Difficulty Level Weighted Loss

July 4, 2022

Classification of Alzheimer's Disease Using the Convolutional Neural Network (CNN) with Transfer Learning and Weighted Loss
Muhammad Wildan Oktavian, Novanto Yudistira, Achmad Ridok
Deep Learning Convolutional Neural Network Transfer Learning Classification Code Supervised ImageNet Alzheimer'S Disease Image Recognition Weighted Loss

March 4, 2022

Deep Learning Neural Networks for Emotion Classification from Text: Enhanced Leaky Rectified Linear Unit Activation and Weighted Loss
Hui Yang, Abeer Alsadoon, P. W. C. Prasad, Thair Al-Dala'in, Tarik A. Rashid, Angelika Maag, Omar Hisham Alsadoon
Deep Neural Network Emotion Classification Improved Classification Sentiment Word Weighted Loss Text Based Emotion Detection

February 4, 2022

Color Image Inpainting via Robust Pure Quaternion Matrix Completion: Error Bound and Weighted Loss
Junren Chen, Michael K. Ng
Error Bound Color Image Completion Method Minimization Problem Weighted Loss Quaternion Matrix Quaternion Matrix Completion

December 13, 2021

Generalization Bounded Implicit Learning of Nearly Discontinuous Functions
Bibit Bianchini, Mathew Halm, Nikolai Matni, Michael Posa
Generalization Bound Generalization Error Bound Weighted Loss Implicit Learning Semi Implicit Implicit Formulation Discontinuous Function

November 15, 2021

Margin-Independent Online Multiclass Learning via Convex Geometry
Guru Guruganesh, Allen Liu, Jon Schneider, Joshua Wang
Binary Classification Convex Set Multi Class Classification Multiclass Classification Weighted Loss Online Multiclass

Weighted Loss

Papers

Robust Time Series Forecasting with Non-Heavy-Tailed Gaussian Loss-Weighted Sampler

Depth Estimation using Weighted-loss and Transfer Learning

MiLe Loss: a New Loss for Mitigating the Bias of Learning Difficulties in Generative Language Models

Tackling Interference Induced by Data Training Loops in A/B Tests: A Weighted Training Approach

MuLMINet: Multi-Layer Multi-Input Transformer Network with Weighted Loss

A Sea-Land Clutter Classification Framework for Over-the-Horizon-Radar Based on Weighted Loss Semi-supervised GAN

A Generalized Weighted Loss for SVC and MLP

Class-Imbalanced Complementary-Label Learning via Weighted Loss

Difficulty-Net: Learning to Predict Difficulty for Long-Tailed Recognition

Classification of Alzheimer's Disease Using the Convolutional Neural Network (CNN) with Transfer Learning and Weighted Loss

Deep Learning Neural Networks for Emotion Classification from Text: Enhanced Leaky Rectified Linear Unit Activation and Weighted Loss

Color Image Inpainting via Robust Pure Quaternion Matrix Completion: Error Bound and Weighted Loss

Generalization Bounded Implicit Learning of Nearly Discontinuous Functions

Margin-Independent Online Multiclass Learning via Convex Geometry