The Thesis Review [04] Sebastian Nowozin - Learning with Structured Data: Applications to Computer Vision
Jul 31, 2020
Sebastian Nowozin, a researcher at Microsoft Research Cambridge, delves into the fascinating world of probabilistic deep learning and its ties to computer vision. He discusses the significance of structured data and innovative loss functions in image analysis. Listeners will learn about the evolution from traditional methods in object recognition to advanced, machine learning-driven techniques. The conversation also covers the challenges and insights in Bayesian deep learning, highlighting the importance of stability in models and sound programming practices in the field.
AI Snips
Chapters
Books
Transcript
Episode notes
Mathematics Structures Research
- Mathematics structures exploration and communication in machine learning research.
- It allows abstraction across problem levels and aids systematic approach discovery.
Unexpected PhD Journey
- Sebastian started programming at nine and had limited math education prior to his PhD.
- He entered a top ML group after a surprise interview talk and overcame imposter syndrome by hard work.
Defining Structured Data
- Structured data means inputs or outputs with elements that have relational dependencies.
- Structured prediction involves mapping inputs to complex output spaces using structured loss functions.
