SiloFuse: Cross-silo Synthetic Data Generation with Latent Tabular Diffusion Models
Privacy-preserving synthesis of feature-partitioned data
Federated Learning
In the realm of data privacy and security, Federated Learning emerges as a cornerstone of Infinidata Lab's approach. This paradigm allows us to train AI models across multiple decentralized devices or servers holding local data samples, without exchanging them. This method ensures that sensitive information remains within its original context, safeguarding privacy while still benefiting from collective insights. Our commitment to Federated Learning exemplifies our dedication to advancing AI technology in a manner that respects and protects user data integrity and confidentiality.
Rihan Hai
I am an assistant professor at Web Information Systems (WIS) group of TU Delft. Before joining TU Delft, I received my Ph.D. degree with “summa cum laude” from Information Systems & Databases group of RWTH Aachen University, Germany. I received my BEng and MEng from Tsinghua University, China. My research focuses on data lakes, data management for machine learning, and quantum data management.
Danning Zhan
PhD candidate working on privacy preservation in machine learning on tabular data in the data lake setting.
Aditya Shankar
I am a PhD candidate at Distributed Systems and Web Information Systems. My supervisors are Dr. Lydia Y. Chen, Dr. Rihan Hai, and Dr. Arie van Deursen. My research interests are distributed learning, privacy-preserving learning, and federated learning.
Find out more content in our Blog
View all posts»The digestion of our on-going projects