Federated Neural Nonparametric Point Processes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Hui, Fan, Xuhui, Liu, Hengyu, Li, Yaqiong, Zhao, Zhilin, Zhou, Feng, Quinn, Christopher John, Cao, Longbing
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914249280847872
author Chen, Hui
Fan, Xuhui
Liu, Hengyu
Li, Yaqiong
Zhao, Zhilin
Zhou, Feng
Quinn, Christopher John
Cao, Longbing
author_facet Chen, Hui
Fan, Xuhui
Liu, Hengyu
Li, Yaqiong
Zhao, Zhilin
Zhou, Feng
Quinn, Christopher John
Cao, Longbing
contents Temporal point processes (TPPs) are effective for modeling event occurrences over time, but they struggle with sparse and uncertain events in federated systems, where privacy is a major concern. To address this, we propose \textit{FedPP}, a Federated neural nonparametric Point Process model. FedPP integrates neural embeddings into Sigmoidal Gaussian Cox Processes (SGCPs) on the client side, which is a flexible and expressive class of TPPs, allowing it to generate highly flexible intensity functions that capture client-specific event dynamics and uncertainties while efficiently summarizing historical records. For global aggregation, FedPP introduces a divergence-based mechanism that communicates the distributions of SGCPs' kernel hyperparameters between the server and clients, while keeping client-specific parameters local to ensure privacy and personalization. FedPP effectively captures event uncertainty and sparsity, and extensive experiments demonstrate its superior performance in federated settings, particularly with KL divergence and Wasserstein distance-based global aggregation.
format Preprint
id arxiv_https___arxiv_org_abs_2410_05637
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Federated Neural Nonparametric Point Processes
Chen, Hui
Fan, Xuhui
Liu, Hengyu
Li, Yaqiong
Zhao, Zhilin
Zhou, Feng
Quinn, Christopher John
Cao, Longbing
Machine Learning
Artificial Intelligence
Cryptography and Security
Temporal point processes (TPPs) are effective for modeling event occurrences over time, but they struggle with sparse and uncertain events in federated systems, where privacy is a major concern. To address this, we propose \textit{FedPP}, a Federated neural nonparametric Point Process model. FedPP integrates neural embeddings into Sigmoidal Gaussian Cox Processes (SGCPs) on the client side, which is a flexible and expressive class of TPPs, allowing it to generate highly flexible intensity functions that capture client-specific event dynamics and uncertainties while efficiently summarizing historical records. For global aggregation, FedPP introduces a divergence-based mechanism that communicates the distributions of SGCPs' kernel hyperparameters between the server and clients, while keeping client-specific parameters local to ensure privacy and personalization. FedPP effectively captures event uncertainty and sparsity, and extensive experiments demonstrate its superior performance in federated settings, particularly with KL divergence and Wasserstein distance-based global aggregation.
title Federated Neural Nonparametric Point Processes
topic Machine Learning
Artificial Intelligence
Cryptography and Security
url https://arxiv.org/abs/2410.05637