Personalized Federated Learning via Gaussian Generative Modeling
Gaussian generative modeling of client heterogeneity with dual-scale fusion for personalized federated learning.
胡马
GeoBrain · People
Peng Hu
Random Noise Denoising Task Lead
Leads random-noise suppression and works on robust noisy-label learning and personalized federated learning.
Random noiseRobust learningFederated learning
Project involvement
Project involvement
Ongoing
Nanjing Institute Prestack Signal-Processing Foundation Model
Signal processing · Foundation models
View project→Ongoing
BGP Prestack Signal-Processing Foundation Model
Signal processing · Foundation models
View project→Ongoing
Multimodal General Foundation Model
Multimodal poststack signal processing
View project→Completed
Loess Plateau Denoising
Ground-roll attenuation
View project→Completed
Nanjing Seismic Deblending
Seismic deblending
View project→Verified work
Verified publications
Robust Losses from Univariate Base Functions for Noisy-Label Learning
A systematic framework for constructing analyzable multiclass robust losses from univariate base functions under label noise.
胡马
Interpolation-Inspired Self-Supervised Method for Seismic Deblending
An interpolation-inspired self-supervised formulation for seismic deblending that learns interference-attenuation representations without complete labels.
胡马