Svitlana Vyetrenko, holder of a Gutenberg Chair, welcomed to BETA

Dr. Svitlana Vyetrenko is joining BETA as the Gutenberg Chair to work on interpretable time series foundation models. Large language models have demonstrated remarkable reasoning and few-shot learning capabilities, and their potential in structured data tasks, such as time series analysis, is emerging. However, challenges persist in identifying patterns and anomalies within time series data and providing contextually meaningful explanations for real-world applications. For example, identifying and explaining suspicious financial transactions while reducing false positives is crucial for loss prevention. Similarly, recognizing and understanding causal relationships is essential for accurate supply chain forecasting. Svitlana’s Gutenberg project will explore novel approaches to constructing interpretable foundational models for time series data, ensuring that AI can capture and explain complex dynamics.

Dr. Svitlana Vyetrenko was previously the AI Research Director at JP Morgan AI Research, as well as a lecturer at the University of California, Berkeley, and Stanford University. Svitlana’s research lies at the intersection of AI, economics, and finance. She is particularly interested in using generative AI to enable more effective decision-making in financial and economic systems. Svitlana holds a PhD in Applied and Computational Mathematics from the California Institute of Technology. She is the author of over 40 peer-reviewed publications and has filed 16 patents. Her paper on multimodal time series retrieval through latent space projections won the Best Industry Paper Award at the International Conference on AI in Finance (ICAIF) in 2023.