S25 GIRS Seminars: One-Step-Ahead Forecasting With Recurrent Variational Quantum Machine Learning

When: Thursday, June 05, 2025
Time: 1:00pm PT
Where: (In-Person) UCLA Engineering 6 BLDG, Rm 580B and ZOOM

Auguste Hirth

Abstract: Currently, existing quantum machine learning architectures, at scales that can be run with near-term quantum hardware, aren’t able to effectively perform forecasting for complex “real” data tasks. This presentation will be a draft research proposal for future work in the evaluation of existing models and proposed improvements, to answer “How far away are useful quantum forecasting models, and how can we get there sooner?”

Short Bio: Auguste Hirth is a PhD student working on quantum machine learning methods designed for time-series problems, with applications in forecasting and prognostics. His current focus is on recurrent models that allow the integration of more data with smaller quantum system sizes, targeting useful applications in the near-term intermediate-scale quantum era.