Summary of Fedvae: Trajectory Privacy Preserving Based on Federated Variational Autoencoder, by Yuchen Jiang et al.
FedVAE: Trajectory privacy preserving based on Federated Variational AutoEncoder
by Yuchen Jiang, Ying Wu, Shiyao Zhang, James J.Q. Yu
First submitted to arxiv on: 12 Jul 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A Federated Variational AutoEncoder (FedVAE) approach is proposed to address the challenge of preserving user confidentiality while utilizing trajectory data in Intelligent Transport Systems (ITS) and Location-Based Services (LBS). The existing methods, such as K-anonymity and Differential Privacy, can introduce perturbations or generate unrealistic data, impacting performance. FedVAE combines Variational AutoEncoder (VAE) to maintain the original feature space, generates new trajectory data, and incorporates Federated Learning (FL) for local data storage, ensuring privacy protection. The results show superior performance compared to existing methods, highlighting FedVAE as a promising solution for enhancing data privacy and utility in location-based applications. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary FedVAE is a new way to keep user information private while using trajectory data in traffic systems. This data has sensitive information about people’s movements and habits, so we need to protect it. Other methods try to do this, but they can make the data not realistic or change its features, which makes them less useful. FedVAE uses a different method that combines two techniques: Variational AutoEncoder (VAE) to keep the original data features and Federated Learning (FL) to store data locally on users’ devices, keeping it private. The results show that FedVAE works better than other methods. |
Keywords
» Artificial intelligence » Federated learning » Variational autoencoder
