Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2404.03702 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929303655022592 |
|---|---|
| author | Liu, Qingxiang Sun, Sheng Liang, Yuxuan Xue, Jingjing Liu, Min |
| author_facet | Liu, Qingxiang Sun, Sheng Liang, Yuxuan Xue, Jingjing Liu, Min |
| contents | The existing federated learning (FL) methods for spatio-temporal forecasting fail to capture the inherent spatio-temporal heterogeneity, which calls for personalized FL (PFL) methods to model the spatio-temporally variant patterns. While contrastive learning approach is promising in addressing spatio-temporal heterogeneity, the existing methods are noneffective in determining negative pairs and can hardly apply to PFL paradigm. To tackle this limitation, we propose a novel PFL method, named Federated dUal sEmantic aLignment-based contraStive learning (FUELS), which can adaptively align positive and negative pairs based on semantic similarity, thereby injecting precise spatio-temporal heterogeneity into the latent representation space by auxiliary contrastive tasks. From temporal perspective, a hard negative filtering module is introduced to dynamically align heterogeneous temporal representations for the supplemented intra-client contrastive task. From spatial perspective, we design lightweight-but-efficient prototypes as client-level semantic representations, based on which the server evaluates spatial similarity and yields client-customized global prototypes for the supplemented inter-client contrastive task. Extensive experiments demonstrate that FUELS outperforms state-of-the-art methods, with communication cost decreasing by around 94%. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_03702 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Personalized Federated Learning for Spatio-Temporal Forecasting: A Dual Semantic Alignment-Based Contrastive Approach Liu, Qingxiang Sun, Sheng Liang, Yuxuan Xue, Jingjing Liu, Min Machine Learning Artificial Intelligence The existing federated learning (FL) methods for spatio-temporal forecasting fail to capture the inherent spatio-temporal heterogeneity, which calls for personalized FL (PFL) methods to model the spatio-temporally variant patterns. While contrastive learning approach is promising in addressing spatio-temporal heterogeneity, the existing methods are noneffective in determining negative pairs and can hardly apply to PFL paradigm. To tackle this limitation, we propose a novel PFL method, named Federated dUal sEmantic aLignment-based contraStive learning (FUELS), which can adaptively align positive and negative pairs based on semantic similarity, thereby injecting precise spatio-temporal heterogeneity into the latent representation space by auxiliary contrastive tasks. From temporal perspective, a hard negative filtering module is introduced to dynamically align heterogeneous temporal representations for the supplemented intra-client contrastive task. From spatial perspective, we design lightweight-but-efficient prototypes as client-level semantic representations, based on which the server evaluates spatial similarity and yields client-customized global prototypes for the supplemented inter-client contrastive task. Extensive experiments demonstrate that FUELS outperforms state-of-the-art methods, with communication cost decreasing by around 94%. |
| title | Personalized Federated Learning for Spatio-Temporal Forecasting: A Dual Semantic Alignment-Based Contrastive Approach |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2404.03702 |