I am a professor at University of Neuchatel and TU Delft. I am the director of Distributed Learning Systems Lab. I returned to academia, after a decade of industry experience at the IBM Research Zurich Lab. My research interests lie in the distinct areas of deep learning, distributed systems, and trust worthy technology. My research is supported by the Swiss National Science Foundation, Dutch National Science Foundation, the European Union, IBM Research, ABB, TU Delft, ASM, Tata Steel, and ASML. My recent research focuses on following research questions.

  • Generative AI systems: how to discover the unnkown via the power of generative models, ranging from tables, time series to graph? I am exploring large langague models and diffusion models in collaboration with scientiest in natural science and manufactoring.
  • Privacy-preserving learning systems by synthetic data and federated learning : how to maximize the knowledge of while maintaining data privacy ? I am combining the deep generative models, federated learning and privacy-enhancing technologies as a privacy-preserving data sharing solution.
  • Trustworthy learning systems: how to make learning algorithms robust and verifiable against adversaries that maliciously manipulate data input? I am designing practical strategies and theories against adversaries, e.g., watermarks


  • News

  • [04/25]Our group has three papers accepted at ECML/PKDD 25
  • [04/25] Lydia will serve DSN 26 TPC co-chairs
  • [04/25] Our group has three papers accepted at ARES 25
  • [02/25] Our paper, “GTV: Generating Tabular Data via Vertical Federated Learning" , is accepted at DSN 25
  • [01/25] Our paper, “TabWak: A Watermark for Tabular Diffusion Models" , is accepted as a spotlight paper at ICLR 25
  • [01/25] Our paper, “CCBNet: Confidential Collaborative Bayesian Networks Inference" , is accepted at Financial Cypto 25
  • [01/25] Our paper, “TS-Inverse: A Gradient Inversion Attack tailored for Federated Time Series Forecasting Models" , is accepted at SatML 25
  • [01/25] Our paper, “TabuLa: Harnessing Language Models for Tabular Data Synthesis" , is accepted at PAKDD 25
  • [08/24] Lydia will serve Middleware 25 TPC co-chairs

  • Old News.....