Feature Alignment

Feature alignment in machine learning aims to harmonize disparate data representations from different sources (e.g., images, text, sensor readings) to improve model performance and generalization. Current research focuses on developing novel alignment strategies within various model architectures, including transformers and graph neural networks, often employing techniques like optimal transport, contrastive learning, and adversarial training to bridge domain gaps. These advancements are crucial for improving the accuracy and robustness of numerous applications, such as medical image analysis, speech recognition, and cross-modal retrieval, where integrating diverse data types is essential. The ultimate goal is to create more powerful and generalizable models capable of handling complex, real-world data.

Papers

July 24, 2023

FaFCNN: A General Disease Classification Framework Based on Feature Fusion Neural Networks
Menglin Kong, Shaojie Zhao, Juan Cheng, Xingquan Li, Ri Su, Muzhou Hou, Cong Cao
Adversarial Learning Feature Fusion Feature Alignment Disease Classification Feature Correlation

July 19, 2023

Revisiting invariances and introducing priors in Gromov-Wasserstein distances
Pinar Demetci, Quang Huy Tran, Ievgen Redko, Ritambhara Singh
Optimal Transport Feature Alignment Alpha Invariance SAM Prior Gromov Wasserstein Gromov Wasserstein Distance Single Cell Multi Omics

July 12, 2023

UGCANet: A Unified Global Context-Aware Transformer-based Network with Feature Alignment for Endoscopic Image Analysis
Pham Vu Hung, Nguyen Duy Manh, Nguyen Thi Oanh, Nguyen Thi Thuy, Dinh Viet Sang
Unified Alignment Feature Alignment Endoscopic Image Transformer Based Deep Context Aware Transformer Endoscopic Imaging

June 29, 2023

Alternative Telescopic Displacement: An Efficient Multimodal Alignment Method
Jiahao Qin, Yitao Xu, Zong Lu, Xiaojun Zhang
Multimodal Data Multimodal Phenomenon Multimodal Information Feature Alignment Mutual Guidance

June 20, 2023

Personalized Federated Learning with Feature Alignment and Classifier Collaboration
Jian Xu, Xinyi Tong, Shao-Lun Huang
Personalized Federated Learning Feature Alignment Global Semantic Classification Head

June 15, 2023

Contrast, Stylize and Adapt: Unsupervised Contrastive Learning Framework for Domain Adaptive Semantic Segmentation
Tianyu Li, Subhankar Roy, Huayi Zhou, Hongtao Lu, Stephane Lathuiliere
Semantic Segmentation Feature Alignment Domain Adaptive Semantic Segmentation unsupeRvised Contrastive Learning Identity Stylization Pixel Level Alignment Contrastive Feature Pixel Contrast

May 20, 2023

How Spurious Features Are Memorized: Precise Analysis for Random and NTK Features
Simone Bombari, Marco Mondelli
Deep Learning Model Neural Tangent Kernel Generalization Error Feature Alignment Comprehensive Analysis Random Participation Spurious Feature Spurious Pattern

May 18, 2023

Manifold-Aware Self-Training for Unsupervised Domain Adaptation on Regressing 6D Object Pose
Yichen Zhang, Jiehong Lin, Ke Chen, Zelin Xu, Yaowei Wang, Kui Jia
Domain Adaptation 6D Object Feature Alignment Domain Invariant Representation Manifold Regularization Video Regression

April 24, 2023

Universal Domain Adaptation via Compressive Attention Matching
Didi Zhu, Yincuan Li, Junkun Yuan, Zexi Li, Kun Kuang, Chao Wu
Vision Transformer Feature Alignment Universal Domain Adaptation

April 17, 2023

K-means Clustering Based Feature Consistency Alignment for Label-free Model Evaluation
Shuyu Miao, Lin Zheng, Jingjing Liu, and Hong Jin
Feature Alignment Visual Data Label Free Model Evaluation First Dataset

March 27, 2023

Global Relation Modeling and Refinement for Bottom-Up Human Pose Estimation
Ruoqi Yin, Jianqin Yin
Convolutional Neural Network Feature Alignment Supervised Keypoint Global Reasoning

March 23, 2023

Three ways to improve feature alignment for open vocabulary detection
Relja Arandjelović, Alex Andonian, Arthur Mensch, Olivier J. Hénaff, Jean-Baptiste Alayrac, Andrew Zisserman
Image Text Pair Text Embeddings Feature Alignment New Way Feature Pyramid Network Open Vocabulary Detection Text Alignment Zero Shot Open Vocabulary

March 20, 2023

Feature Alignment and Uniformity for Test Time Adaptation
Shuai Wang, Daoan Zhang, Zipei Yan, Jianguo Zhang, Rui Li
Test Time Adaptation Test Time Feature Alignment Uniformity Metric

March 14, 2023

GANN: Graph Alignment Neural Network for Semi-Supervised Learning
Linxuan Song, Wenxuan Tu, Sihang Zhou, Xinwang Liu, En Zhu
Semi Supervised Learning Semi Supervised Feature Alignment Semi Supervised Node Classification Graph Alignment Optimal Alignment

March 12, 2023

Extending global-local view alignment for self-supervised learning with remote sensing imagery
Xinye Wanyan, Sachith Seneviratne, Shuchang Shen, Michael Kirley
Self Supervised Learning Remote Sensing Image Feature Alignment Visual Imagery Remote Sensing Task State of the Art Self

March 8, 2023

FastFill: Efficient Compatible Model Update
Florian Jaeckle, Fartash Faghri, Ali Farhadi, Oncel Tuzel, Hadi Pouransari
Retrieval Performance Feature Alignment Compatible Learning

February 3, 2023

Class Overwhelms: Mutual Conditional Blended-Target Domain Adaptation
Pengcheng Xu, Boyu Wang, Charles Ling
Feature Alignment True Class Domain Variation Target Domain Adaptation

November 17, 2022

FedFA: Federated Learning with Feature Anchors to Align Features and Classifiers for Heterogeneous Data
Tailin Zhou, Jun Zhang, Danny H. K. Tsang
Simple Classifier Heterogeneous Data Feature Space Feature Alignment Visual Information Anchor Data Divergence

November 13, 2022

Generalization Beyond Feature Alignment: Concept Activation-Guided Contrastive Learning
Yibing Liu, Chris Xing Tian, Haoliang Li, Shiqi Wang
Contrastive Learning Strong Generalization Feature Alignment Invariant Representation Qualitative Difference High Impact Concept Neuron Level

October 26, 2022

Trade-off between reconstruction loss and feature alignment for domain generalization
Thuan Nguyen, Boyang Lyu, Prakash Ishwar, Matthias Scheutz, Shuchin Aeron
Latent Representation Unseen Domain Feature Alignment Domain Alignment Reconstruction Loss

Feature Alignment

Papers

FaFCNN: A General Disease Classification Framework Based on Feature Fusion Neural Networks

Revisiting invariances and introducing priors in Gromov-Wasserstein distances

UGCANet: A Unified Global Context-Aware Transformer-based Network with Feature Alignment for Endoscopic Image Analysis

Alternative Telescopic Displacement: An Efficient Multimodal Alignment Method

Personalized Federated Learning with Feature Alignment and Classifier Collaboration

Contrast, Stylize and Adapt: Unsupervised Contrastive Learning Framework for Domain Adaptive Semantic Segmentation

How Spurious Features Are Memorized: Precise Analysis for Random and NTK Features

Manifold-Aware Self-Training for Unsupervised Domain Adaptation on Regressing 6D Object Pose

Universal Domain Adaptation via Compressive Attention Matching

K-means Clustering Based Feature Consistency Alignment for Label-free Model Evaluation

Global Relation Modeling and Refinement for Bottom-Up Human Pose Estimation

Three ways to improve feature alignment for open vocabulary detection

Feature Alignment and Uniformity for Test Time Adaptation

GANN: Graph Alignment Neural Network for Semi-Supervised Learning

Extending global-local view alignment for self-supervised learning with remote sensing imagery

FastFill: Efficient Compatible Model Update

Class Overwhelms: Mutual Conditional Blended-Target Domain Adaptation

FedFA: Federated Learning with Feature Anchors to Align Features and Classifiers for Heterogeneous Data

Generalization Beyond Feature Alignment: Concept Activation-Guided Contrastive Learning

Trade-off between reconstruction loss and feature alignment for domain generalization