Introduction to deep transfer learning with Xfer
- Date: Friday 8 March 2019, 14:00 – 15:00
- Location: Mathematics Level 8, MALL 1, School of Mathematics
- Type: Seminars, Statistics
- Cost: Free
Andreas Damianou, Amazon, Cambridge. Part of the Statistics Seminar Series.
In this talk I will firstly give an overview of the fundamentals of deep neural networks and offer a complementary view through the lens of probabilistic modeling. For the second part of the talk I will focus on transfer learning with deep neural networks, a set of techniques that allows us to reuse and repurpose already trained models. The relevant discussion will also contain a description of Xfer, a library for deep transfer learning. Relevant background can be found in this blog post and in this introductory notebook.
I am a Machine Learning scientist at Amazon, Cambridge. My research focuses on Bayesian probabilistic modeling as well as on methods for improving deep learning through knowledge transfer and uncertainty quantification. I'm particularly interested in applying my research in decision making systems. In the past I have worked in perceptual models for bio-inspired robotics. I have pursued a PhD degree with Neil Lawrence at Sheffield, developing Deep Gaussian Processes.