Metric Learning Loss

Metric learning loss functions aim to learn embedding spaces where semantically similar data points are clustered closely together, while dissimilar points are separated. Current research focuses on developing novel loss functions that improve efficiency, accuracy, and robustness, particularly within Siamese networks and for multimodal data, often employing techniques like triplet loss, proxy losses, and Wasserstein distance. These advancements are impacting various fields, including image retrieval, object detection, and medical image analysis, by enabling more accurate and efficient similarity comparisons and improved model performance in challenging scenarios like few-shot learning and out-of-distribution detection.

Papers

June 8, 2024

Relational Proxy Loss for Audio-Text based Keyword Spotting
Youngmoon Jung, Seungjin Lee, Joon-Young Yang, Jaeyoung Roh, Chang Woo Han, Hoon-Young Cho
Text Embeddings Keyword Spotting Acoustic Word Embeddings Keyword Enrollment Metric Learning Loss Proxy Based Loss

May 1, 2024

Deep Metric Learning-Based Out-of-Distribution Detection with Synthetic Outlier Exposure
Assefa Seyoum Wahd
Denoising Diffusion Probabilistic Model Metric Learning Deep Metric Learning Metric Learning Loss Based Out of Distribution Detection

January 21, 2024

Squared Wasserstein-2 Distance for Efficient Reconstruction of Stochastic Differential Equations
Mingtao Xia, Xiangting Li, Qijing Shen, Tom Chou
Wasserstein Distance Noisy Data Stochastic Differential Equation Numerical Experiment Efficient Reconstruction Metric Learning Loss

November 23, 2023

Shadow: A Novel Loss Function for Efficient Training in Siamese Networks
Alif Elham Khan, Mohammad Junayed Hasan, Humayra Anjum, Nabeel Mohammed
Loss Function Siamese Network Efficient Training Quantum Shadow Triplet Loss Metric Learning Loss

August 21, 2023

Deep Metric Loss for Multimodal Learning
Sehwan Moon, Hyunju Lee
Multimodal Phenomenon Multimodal Model Multimodal Learning Multimodal Dataset Multimodal Feature Metric Learning Loss

June 4, 2023

Active Inference-Based Optimization of Discriminative Neural Network Classifiers
Faezeh Fallah
Objective Function Generating User Activity Metric Learning Loss

April 19, 2023

Improving Post-Training Quantization on Object Detection with Task Loss-Guided Lp Metric
Lin Niu, Jiawei Liu, Zhihang Yuan, Dawei Yang, Xinggang Wang, Wenyu Liu
Object Detection Post Training Quantization Metric Learning Loss Detection Loss Quantization Performance

April 16, 2023

Learning Empirical Bregman Divergence for Uncertain Distance Representation
Zhiyuan Li, Ziru Liu, Anna Zou, Anca L. Ralescu
Deep Metric Learning Bregman Divergence Metric Learning Loss

February 8, 2023

Local Law 144: A Critical Analysis of Regression Metrics
Giulio Filippi, Sara Zannone, Airlie Hilliard, Adriano Koshiyama
Critical Review Bias Metric Metric Learning Loss Bias Auditing Local Rule

December 29, 2022

HIER: Metric Learning Beyond Class Labels via Hierarchical Regularization
Sungyeon Kim, Boseung Jeong, Suha Kwak
Metric Learning Metric Learning Loss Class Label Grained Supervision Latent Hierarchy Hyperbolic Metric Learning Hierarchical Regularisation

December 28, 2022

Joint Discriminative and Metric Embedding Learning for Person Re-Identification
Sinan Sabri, Zaigham Randhawa, Gianfranco Doretto
LeArning Abstract Person Re Identification North Star Metric Discriminative Task Metric Learning Loss Softmax Loss Discriminative Sample Aware Loss Function

November 29, 2022

Intra-class Adaptive Augmentation with Neighbor Correction for Deep Metric Learning
Zheren Fu, Zhendong Mao, Bo Hu, An-An Liu, Yongdong Zhang
Deep Metric Learning Metric Learning Loss Intra Class Variation Adaptive Augmentation Order Neighbor

November 2, 2022

Rethinking the Metric in Few-shot Learning: From an Adaptive Multi-Distance Perspective
Jinxiang Lai, Siqian Yang, Guannan Jiang, Xi Wang, Yuxi Li, Zihui Jia, Xiaochen Chen, Jun Liu, Bin-Bin Gao, Wei Zhang, Yuan Xie, Chengjie Wang
Shot Classification North Star Metric Metric Learning Loss Shot Classifier Metric Prediction

October 21, 2022

Dissecting Deep Metric Learning Losses for Image-Text Retrieval
Hong Xuan, Xi Chen
Vision Language Model Deep Metric Learning Image Text Retrieval Metric Learning Loss Gradient Space

August 18, 2022

Siamese Prototypical Contrastive Learning
Shentong Mo, Zhun Sun, Chao Li
Contrastive Learning Contrastive Loss Visual Representation Contrastive Self Supervised Learning Metric Learning Loss Contrastive Siamese

July 12, 2022

Regression Metric Loss: Learning a Semantic Representation Space for Medical Images
Hanqing Chao, Jiajin Zhang, Pingkun Yan
Loss Function Medical Image Metric Learning Loss Regression Loss

June 22, 2022

Feature Re-calibration based Multiple Instance Learning for Whole Slide Image Classification
Philip Chikontwe, Soo Jeong Nam, Heounjeong Go, Meejeong Kim, Hyun Jung Sung, Sang Hyun Park
Multiple Instance Learning Multi Head Self Attention Whole Slide Image Classification Fully Supervised Metric Learning Loss Feature Calibration

May 19, 2022

Bypassing Logits Bias in Online Class-Incremental Learning with a Generative Framework
Gehui Shen, Shibo Jie, Ziheng Li, Zhi-Hong Deng
Continual LEArning Online Continual Learning Generative Framework Memory Replay Second Ranked Logits Metric Learning Loss Online Class Incremental Learning

May 5, 2022

Do Different Deep Metric Learning Losses Lead to Similar Learned Features?
Konstantin Kobs, Michael Steininger, Andrzej Dulny, Andreas Hotho
Loss Function Saliency Map Metric Learning Loss Metric Learning Loss Function

April 14, 2022

Surface Similarity Parameter: A New Machine Learning Loss Metric for Oscillatory Spatio-Temporal Data
Mathies Wedler, Merten Stender, Marco Klein, Svenja Ehlers, Norbert Hoffmann
Loss Function Metric Learning Loss Spatiotemporal Pattern Surface Parameterization

Metric Learning Loss

Papers

Relational Proxy Loss for Audio-Text based Keyword Spotting

Deep Metric Learning-Based Out-of-Distribution Detection with Synthetic Outlier Exposure

Squared Wasserstein-2 Distance for Efficient Reconstruction of Stochastic Differential Equations

Shadow: A Novel Loss Function for Efficient Training in Siamese Networks

Deep Metric Loss for Multimodal Learning

Active Inference-Based Optimization of Discriminative Neural Network Classifiers

Improving Post-Training Quantization on Object Detection with Task Loss-Guided Lp Metric

Learning Empirical Bregman Divergence for Uncertain Distance Representation

Local Law 144: A Critical Analysis of Regression Metrics

HIER: Metric Learning Beyond Class Labels via Hierarchical Regularization

Joint Discriminative and Metric Embedding Learning for Person Re-Identification

Intra-class Adaptive Augmentation with Neighbor Correction for Deep Metric Learning

Rethinking the Metric in Few-shot Learning: From an Adaptive Multi-Distance Perspective

Dissecting Deep Metric Learning Losses for Image-Text Retrieval

Siamese Prototypical Contrastive Learning

Regression Metric Loss: Learning a Semantic Representation Space for Medical Images

Feature Re-calibration based Multiple Instance Learning for Whole Slide Image Classification

Bypassing Logits Bias in Online Class-Incremental Learning with a Generative Framework

Do Different Deep Metric Learning Losses Lead to Similar Learned Features?

Surface Similarity Parameter: A New Machine Learning Loss Metric for Oscillatory Spatio-Temporal Data