Of all machine learning methods, generative models are particularly interesting for scientific applications because of their probabilistic nature and ability to fit complex data and probability distributions. However, in their vanilla forms, generative models have a number of shortcomings and failure modes which can be a hindrance to their application: they can be difficult to train on high dimensional data, and they can fail in crucial tasks such as outlier detection or the generation of realistic artificial data. In my talk, I am going to explore the reasons for these failures and propose new generative models and generative model based approaches that are robust to these shortcomings. The proposed approaches are easy to train and validate, numerically stable, and do not require fine-tuning. They should thus be particularly fitted for scientific applications. I will demonstrate how these approaches can be used for scientifically relevant tasks such as realistic data generation and outlier detection.
Speaker: Vanessa Bohm, UC Berkeley
This event is online. Link to Zoom is available at the event weblink.
Contact:Website: Click to Visit
Save this Event:iCalendar
Windows Live Calendar