Research map / Federated learning and privacy / Personalized and heterogeneous FL
FedVS: Straggler-Resilient and Privacy-Preserving Vertical Federated Learning for Split Models
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use a 3-layer CNN instead of a 3-layer PN as the client network | Split VFL with VGG13 server · FashionMNIST | test accuracy: 88.49% -> 91.53% | Helped |
| Use a 3-layer MLP instead of a 3-layer PN as the client network | Split VFL with VGG13 server · FashionMNIST | test accuracy: 88.49% -> 88.70% | Helped |
| Increase the number of PN layers from 1 to 3 (ablation) | Split VFL · FashionMNIST | test accuracy: 88.19% -> 88.49% | Helped |
| Replace element-wise average embedding aggregation with concatenation (ablation) | FedVS · FashionMNIST, no stragglers | test accuracy | No clear effect |
| Choose a smaller partition parameter K (ablation) | FedVS · High-delay straggler pattern | test accuracy | Helped |
| Choose a larger partition parameter K (ablation) | FedVS · Local computation dominates delay | test accuracy | Helped |
| Increase privacy parameter T (ablation) | FedVS · Multiple VFL datasets | test accuracy | Hurt |
| Add calibrated differential-privacy noise to client embeddings | Wait-DP · All evaluated datasets | test accuracy | Hurt |
| Use FedVS rather than the evaluated baseline protocols | Split VFL · CV and multi-view datasets | test accuracy | Helped |
| Use FedVS rather than VAFL or Ignore | Split VFL · Tabular datasets | test accuracy | Helped |
Closest papers by meaning
- Complementary Knowledge Distillation for Robust and Privacy-Preserving Model Serving in Vertical Federated Learning (AAAI 2024 · similarity 0.88)
- PubSub-VFL: Towards Efficient Two-Party Split Learning in Heterogeneous Environments via Publisher/Subscriber Architecture (NeurIPS 2025 · similarity 0.88)
- FedLMT: Tackling System Heterogeneity of Federated Learning via Low-Rank Model Training with Theoretical Guarantees (ICML 2024 · similarity 0.87)
- VFLAIR: A Research Library and Benchmark for Vertical Federated Learning (ICLR 2024 · similarity 0.87)
- Federated Transformer: Multi-Party Vertical Federated Learning on Practical Fuzzily Linked Data (NeurIPS 2024 · similarity 0.87)
- Federated Learning via Input-Output Collaborative Distillation (AAAI 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.