< Back to previous page

Project

Label Efficient AI architecture for Screen-Task WORKflow assistance (LEASTWORK)

The PhD topic and goal of the Baekeland project is a use case-agnostic artificial intelligence (AI) architecture for workflow assistance and automatic logging, that shortens the time-to-market by being reusable and less reliant on expensive, expert annotations. Self-supervised learning and short-term planning in a learned world model will be studied in two or more use cases relevant to the company Barco, including robotic surgery.

Date:16 Sep 2021 →  Today
Keywords:Machine Learning, Deep Learning, Self-supervision, Representation learning, Computer Vision, Surgery, Imitation learning, Learned world model
Disciplines:Machine learning and decision making
Project type:PhD project