Seeing things clearly: Image understanding through hard-attention and reasoning with structured knowledges
University of the Witwatersrand
Abstract
In this talk, Jonathan aims to frame the current challenges of explainability and understanding in ML-driven approaches to image processing, and their potential solution through explicit inference techniques.
Topics
Keep track of Computer Vision
Remember this field on this device and see what changed when you return. Your saved items stay in your library.
The calendar subscription updates upcoming seminars automatically. Add the feed by URL in your calendar; a one-off import will not update. Attendance conditions remain those of the organiser.
Related Seminars
Deep learning applications in ophthalmology
More on artificial intelligence and machine learning
Do large language models solve verbal analogies like children do?
More on artificial intelligence and reasoning
Zero-shot visual reasoning with probabilistic analogical mapping
More on artificial intelligence and reasoning