PhD lunch lecture
June 10 @ 12:30 – 13:30
Are you interested in using AI for medical images, computer vision, or maybe even doing a PhD? In this lunch lecture, two old AI&ES master students and current PhD’ers will tell you more about this.
They will give you some information that you can use to decide whether a PhD is something for you. Of course, they will also talk about their own research.
Simon Penninga uses generative AI techniques (diffusion models and VAEs) to create an agent that functions as an “active perceiver” in ultrasound imaging. He uses Bayesian posterior belief to generate images that align with the observations to create adaptive sampling schemes in real time to maximize information gain. Using this he can reach subsampling levels of up to 20-fold and still reproduce cardiac images.
Lemar Abdi does his research in the computer science field. He focuses on unsupervised out-of-distribution (OOD) detection using generative models. He will also show how information-theoretic insights can lead to state-of-the-art performance in OOD detection for medical images.
This lunch lecture will be held on the 10th of June from 12:30 – 13:30. Lunch will be provided.