Unknown Heterogeneity

Unknown heterogeneity, the presence of unobserved variations within datasets, poses a significant challenge across diverse machine learning applications, from federated learning to seismic analysis and medical treatment optimization. Current research focuses on developing algorithms and model architectures that can effectively handle this heterogeneity, including methods leveraging federated learning with adaptive client selection, attention mechanisms in deep neural networks, and techniques that incorporate propensity scores or class-attribute priors to account for varying data distributions. Addressing unknown heterogeneity is crucial for improving the robustness, generalizability, and fairness of machine learning models, leading to more reliable and impactful results in various scientific and practical domains.

Papers

November 7, 2024

Towards Robust Federated Analytics via Differentially Private Measurements of Statistical Heterogeneity
Mary Scott, Graham Cormode, Carsten Maple
Differential Privacy Homogeneity Bias Statistical Heterogeneity Federated Data Internal Measurement Unknown Heterogeneity Privacy Parameter

August 15, 2024

Addressing Skewed Heterogeneity via Federated Prototype Rectification with Personalization
Shunxin Guo, Hongsong Wang, Shuxia Lin, Zhiqiang Kou, Xin Geng
Personalized Federated Learning Heterogeneous Federated Learning Prototype Alignment Unknown Heterogeneity

June 25, 2024

Navigating High-Degree Heterogeneity: Federated Learning in Aerial and Space Networks
Fan Dong, Henry Leung, Steve Drew
Class Imbalance Unobserved Heterogeneity Heterogeneity Challenge Unknown Heterogeneity

April 15, 2024

High-Resolution Detection of Earth Structural Heterogeneities from Seismic Amplitudes using Convolutional Neural Networks with Attention layers
Luiz Schirmer, Guilherme Schardong, Vinícius da Silva, Rogério Santos, Hélio Lopes
Convolutional Neural Network Attention Layer Amplitude Estimation Unknown Heterogeneity Structural Heterogeneity

February 19, 2024

Beyond Uniform Scaling: Exploring Depth Heterogeneity in Neural Architectures
Akash Guna R. T, Arnav Chavan, Deepak Gupta
Vision Transformer Neural Architecture Loss Landscape Multiplicative Size Scaling ImageNet LT Unknown Heterogeneity Uniform Scaling

February 13, 2024

FLASH: Federated Learning Across Simultaneous Heterogeneities
Xiangyu Chang, Sk Miraj Ahmed, Srikanth V. Krishnamurthy, Basak Guler, Ananthram Swami, Samet Oymak, Amit K. Roy-Chowdhury
Client Selection Client Heterogeneity Linear Time Algorithm Unknown Heterogeneity Latency Critical

February 2, 2024

FedShift: Tackling Dual Heterogeneity Problem of Federated Learning via Weight Shift Aggregation
Jungwon Seo, Chunming Rong, Minhoe Kim
Weight Aggregation Heterogeneity Challenge System Heterogeneity Unknown Heterogeneity Heterogeneity Environment

January 25, 2024

Class-attribute Priors: Adapting Optimization to Heterogeneity and Fairness Objective
Xuechen Zhang, Mingchen Li, Jiasi Chen, Christos Thrampoulidis, Samet Oymak
Loss Function Optimization Purpose Optimal Classifier Fairness Objective Class Prior Unknown Heterogeneity Label Imbalance

December 12, 2023

Combining propensity score methods with variational autoencoders for generating synthetic data in presence of latent sub-groups
Kiana Farhadyar, Federico Bonofiglio, Maren Hackenberg, Daniela Zoeller, Harald Binder
Synthetic Data Variational Autoencoders Synthetic Data Generation Speech Presence Propensity Score Unobserved Heterogeneity Unknown Heterogeneity

November 12, 2023

FLASH-RL: Federated Learning Addressing System and Static Heterogeneity using Reinforcement Learning
Sofiane Bouaziz, Hadjer Benmeziane, Youcef Imine, Leila Hamdad, Smail Niar, Hamza Ouarnoughi
Reinforcement Learning Mobile User Unknown Heterogeneity Double Deep Q Learning

October 4, 2023

Sweeping Heterogeneity with Smart MoPs: Mixture of Prompts for LLM Task Adaptation
Chen Dun, Mirian Hipolito Garcia, Guoqing Zheng, Ahmed Hassan Awadallah, Anastasios Kyrillidis, Robert Sim
Mixture Component Complex Prompt Prompt Tuning Task Specification LLM Adaptation Cleaning Robot Heterogeneous Task Unknown Heterogeneity

November 28, 2022

Meta-analysis of individualized treatment rules via sign-coherency
Jay Jojo Cheng, Jared D. Huling, Guanhua Chen
Meta Analysis Model Generalizability Semantic Coherence Unknown Heterogeneity

October 22, 2022

Adaptive Data Fusion for Multi-task Non-smooth Optimization
Henry Lam, Kaizheng Wang, Yuhang Wu, Yichen Zhang
Statistical Learning Statistical Guarantee Data Fusion Non Smooth Unknown Heterogeneity Adaptive Space Fusion

March 11, 2022

ZIN: When and How to Learn Invariance Without Environment Partition?
Yong Lin, Shengyu Zhu, Lu Tan, Peng Cui
Invariant Representation Alpha Invariance Invariant Feature Invariant Model Unknown Heterogeneity Invariant Learning Method

Unknown Heterogeneity

Papers

Towards Robust Federated Analytics via Differentially Private Measurements of Statistical Heterogeneity

Addressing Skewed Heterogeneity via Federated Prototype Rectification with Personalization

Navigating High-Degree Heterogeneity: Federated Learning in Aerial and Space Networks

High-Resolution Detection of Earth Structural Heterogeneities from Seismic Amplitudes using Convolutional Neural Networks with Attention layers

Beyond Uniform Scaling: Exploring Depth Heterogeneity in Neural Architectures

FLASH: Federated Learning Across Simultaneous Heterogeneities

FedShift: Tackling Dual Heterogeneity Problem of Federated Learning via Weight Shift Aggregation

Class-attribute Priors: Adapting Optimization to Heterogeneity and Fairness Objective

Combining propensity score methods with variational autoencoders for generating synthetic data in presence of latent sub-groups

FLASH-RL: Federated Learning Addressing System and Static Heterogeneity using Reinforcement Learning

Sweeping Heterogeneity with Smart MoPs: Mixture of Prompts for LLM Task Adaptation

Meta-analysis of individualized treatment rules via sign-coherency

Adaptive Data Fusion for Multi-task Non-smooth Optimization

ZIN: When and How to Learn Invariance Without Environment Partition?