dorsal/arxiv
View SchemaPiXTime: A Model for Federated Time Series Forecasting with Heterogeneous Data Structures Across Nodes
| Authors | Yiming Zhou, Mingyue Cheng, Hao Wang, Enhong Chen |
|---|---|
| Categories | |
| ArXiv ID | 2601.05613vv1 |
| URL | https://arxiv.org/abs/2601.05613 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
Time series are highly valuable and rarely shareable across nodes, making federated learning a promising paradigm to leverage distributed temporal data. However, different sampling standards lead to diverse time granularities and variable sets across nodes, hindering classical federated learning. We propose PiXTime, a novel time series forecasting model designed for federated learning that enables effective prediction across nodes with multi-granularity and heterogeneous variable sets. PiXTime employs a personalized Patch Embedding to map node-specific granularity time series into token sequences of a unified dimension for processing by a subsequent shared model, and uses a global VE Table to align variable category semantics across nodes, thereby enhancing cross-node transferability. With a transformer-based shared model, PiXTime captures representations of auxiliary series with arbitrary numbers of variables and uses cross-attention to enhance the prediction of the target series. Experiments show PiXTime achieves state-of-the-art performance in federated settings and demonstrates superior performance on eight widely used real-world traditional benchmarks.
{
"annotation_id": "396f9aed-eb7e-4cc2-aeab-9764f5d5dadb",
"date_created": "2026-02-17T05:53:05.118000Z",
"date_modified": "2026-02-17T05:53:05.118000Z",
"file_hash": "6ffeff5e12c495a4d7224720ed4df2eb821e1b1de5e7223161a3f1b05465f42e",
"private": false,
"record": {
"abstract": "Time series are highly valuable and rarely shareable across nodes, making federated learning a promising paradigm to leverage distributed temporal data. However, different sampling standards lead to diverse time granularities and variable sets across nodes, hindering classical federated learning. We propose PiXTime, a novel time series forecasting model designed for federated learning that enables effective prediction across nodes with multi-granularity and heterogeneous variable sets. PiXTime employs a personalized Patch Embedding to map node-specific granularity time series into token sequences of a unified dimension for processing by a subsequent shared model, and uses a global VE Table to align variable category semantics across nodes, thereby enhancing cross-node transferability. With a transformer-based shared model, PiXTime captures representations of auxiliary series with arbitrary numbers of variables and uses cross-attention to enhance the prediction of the target series. Experiments show PiXTime achieves state-of-the-art performance in federated settings and demonstrates superior performance on eight widely used real-world traditional benchmarks.",
"arxiv_id": "2601.05613",
"authors": [
"Yiming Zhou",
"Mingyue Cheng",
"Hao Wang",
"Enhong Chen"
],
"categories": [
"cs.LG",
"cs.AI"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "PiXTime: A Model for Federated Time Series Forecasting with Heterogeneous Data Structures Across Nodes",
"url": "https://arxiv.org/abs/2601.05613",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "25a2ecdd-1bd2-4adb-844a-a3b4859b6f21",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}