Delft AI Lab Lunch with BIO Lab and DI_Lab

About BIO Lab

Biomedical Intervention Optimisation

Modern machine learning algorithms have achieved unprecedented accuracy in image and video understanding tasks, via pure learning from data. These powerful abilities come at the price of enormous amounts of training data, memory, and computational requirements. Such resources are rarely available to real-time feedback systems in medical intervention and biomedical research.

Experts will join forces in the BIOLab across many fields including computer vision, reinforcement learning, neural architecture, deep learning and computational physics, and biomedical imaging. We will create high-efficiency, real-time, AI-driven feedback and control in biomedical applications. The focus is on improving the efficiency of machine learning algorithms by designing novel artificial neural network architectures, developing new reinforcement learning and generative algorithms, and incorporating biologically inspired neural network models. These newly developed concepts and algorithms will be applied to a wide range of problems in biomedical applications. Examples include optimizing tumour irradiation protocols with missing information, and limiting irradiation damage to delicate living samples in smart microscopy.  

About DI_Lab

Humans and AI designing the future in dialogue

At the Designing Intelligence Lab, our overarching goal is to identify and design the conditions, methods, and tools for human and artificial intelligence to creatively work together. We split this into the following objectives:

• Multimodal Processes. We are pioneering multimodal, multi-party frameworks for data collection and training of new machine learning algorithms to develop real-time analytic methods for social robotics.
• Hybrid Intelligence. We are creating virtual agents and robots that can meaningfully interact and collaborate with humans over extended periods of time, in a variety of design tasks.
• Creative Modalities. Using big data, we are generating models correlating design expertise with agency, allowing us to experiment with artificial agency during complex system design processes.
• Next-Gen Design Methods. We are exploring the form and use of novel design methods to address systemic design problems to create an AI Toolkit.