Class Diversity

Class diversity, encompassing both intra-class (variations within a category) and inter-class (differences between categories) variability, is a crucial factor influencing the performance of machine learning models, particularly in image classification. Current research focuses on improving model generalization to unseen classes by leveraging intra-class diversity through techniques like soft labeling during pre-training, representing classes with multiple vectors instead of a single one, and incorporating diverse synthetic data generation. Addressing class imbalance and effectively utilizing this diversity is vital for enhancing model accuracy, robustness, and interpretability across various applications, including biomedical image analysis and zero-shot learning.

Papers

October 26, 2024

Anatomical 3D Style Transfer Enabling Efficient Federated Learning with Extremely Low Communication Costs
Yuto Shibata, Yasunori Kudo, Yohei Sugawara
Multi Organ Segmentation Communication Cost 3D Style Transfer Multi Organ Class Diversity

October 14, 2024

Class Balancing Diversity Multimodal Ensemble for Alzheimer's Disease Diagnosis and Early Detection
Arianna Francesconi, Lazzaro di Biase, Donato Cappetta, Fabio Rebecchi, Paolo Soda, Rosa Sicilia, Valerio Guarrasi
Alzheimer'S Disease Class Imbalance Medical Diagnosis Common Disease Early Detection Imbalanced Data Class Diversity

July 2, 2024

Unleash the Power of Local Representations for Few-Shot Classification
Shi Tang, Guiming Luo, Xinchen Ye, Zhiyi Xia
Classification Code Real Power Shot Classification Local Feature Local Representation Class Diversity Novel Class Generalization

April 25, 2024

Embracing Diversity: Interpretable Zero-shot classification beyond one vector per class
Mazda Moayeri, Michael Rabbat, Mark Ibrahim, Diane Bouchacourt
Zero Shot Diversity Awareness True Class Zero Shot Classification X Vector Object State Class Diversity

July 23, 2023

Assessing Intra-class Diversity and Quality of Synthetically Generated Images in a Biomedical and Non-biomedical Setting
Muhammad Muneeb Saad, Mubashir Husain Rehmani, Ruairi O'Reilly
Synthetic Image Large Scale Synthetic Biomedical Research Biomedical Image Non Expert Imaging Modality Biomedical Image Analysis Class Diversity

May 20, 2023

On the Trade-off of Intra-/Inter-class Diversity for Supervised Pre-training
Jieyu Zhang, Bohan Wang, Zhengyu Hu, Pang Wei Koh, Alexander Ratner
Pre Training Supervised Pre Training Class Diversity

March 22, 2023

Re-thinking Federated Active Learning based on Inter-class Diversity
SangMook Kim, Sangmin Bae, Hwanjun Song, Se-Young Yun
Active Learning Local Model Selection Policy Class Diversity Federated Active Learning

October 17, 2022

Understanding CNN Fragility When Learning With Imbalanced Data
Damien Dablain, Kristen N. Jacobson, Colin Bellinger, Mark Roberts, Nitesh Chawla
Deep Learning Convolutional Neural Network CNN Network Imbalanced Data Latent Class Class Diversity Intrinsic Class

September 7, 2022

Blessing of Class Diversity in Pre-training
Yulai Zhao, Jianshu Chen, Simon S. Du
Natural Language Processing Pre Training Language Modeling Task Pre Training Task Blessing Generation Class Diversity

July 13, 2022

D-CBRS: Accounting For Intra-Class Diversity in Continual Learning
Yasin Findik, Farhad Pourkamali-Anaraki
Continual LEArning Continual Learning Replay Based Class Diversity

April 13, 2022

Dynamic Diagnosis of the Progress and Shortcomings of Student Learning using Machine Learning based on Cognitive, Social, and Emotional Features
Alex Doboli, Simona Doboli, Ryan Duke, Sangjin Hong, Wendy Tang
Machine Learning Cognitive Learning Emotional Feature Student Learning Adaptive Teaching Class Diversity Consistency Based Diagnosis

December 15, 2021

Imagine by Reasoning: A Reasoning-Based Implicit Semantic Data Augmentation for Long-Tailed Classification
Xiaohua Chen, Yucan Zhou, Dayan Wu, Wanqian Zhang, Yu Zhou, Bo Li, Weiping Wang
Knowledge Graph Data Augmentation Complex Reasoning ImageNet LT Long Tailed Classification Class Diversity Semantic Data Augmentation

November 17, 2021

IntraQ: Learning Synthetic Images with Intra-Class Heterogeneity for Zero-Shot Network Quantization
Yunshan Zhong, Mingbao Lin, Gongrui Nan, Jianzhuang Liu, Baochang Zhang, Yonghong Tian, Rongrong Ji
Synthetic Image Soft Label Zero Shot Quantization Class Diversity

Class Diversity

Papers

Anatomical 3D Style Transfer Enabling Efficient Federated Learning with Extremely Low Communication Costs

Class Balancing Diversity Multimodal Ensemble for Alzheimer's Disease Diagnosis and Early Detection

Unleash the Power of Local Representations for Few-Shot Classification

Embracing Diversity: Interpretable Zero-shot classification beyond one vector per class

Assessing Intra-class Diversity and Quality of Synthetically Generated Images in a Biomedical and Non-biomedical Setting

On the Trade-off of Intra-/Inter-class Diversity for Supervised Pre-training

Re-thinking Federated Active Learning based on Inter-class Diversity

Understanding CNN Fragility When Learning With Imbalanced Data

Blessing of Class Diversity in Pre-training

D-CBRS: Accounting For Intra-Class Diversity in Continual Learning

Dynamic Diagnosis of the Progress and Shortcomings of Student Learning using Machine Learning based on Cognitive, Social, and Emotional Features

Imagine by Reasoning: A Reasoning-Based Implicit Semantic Data Augmentation for Long-Tailed Classification

IntraQ: Learning Synthetic Images with Intra-Class Heterogeneity for Zero-Shot Network Quantization