Niv Nayman | On Residual Connections
Details
Join Zoom Meeting
https://zoom.us/j/6043600514?pwd=VTFuU2VSTTNhTE1RRFJTZjhZNTN1Zz09
Meeting ID: 604 360 0514
Password: 703769
Abstract:
We got used to the fact that very deep neural networks (DNNs) of hundreds of layers can be trained effectively. Actually, deep models used to be fairly hard to train. In this lecture, we will provide a historical overview of the pursuit after an efficient training of DNNs. In particular, following a series of papers, we will analyse the role of arguably the most important key for training DNNs - residual connections, over their many variants.
- None of the presented works are of the lecturer.
Bio:
Niv Nayman is an algorithm engineer at Alibaba Machine Intelligence Israel Lab, working primarily on AutoML research and application. Before joining Alibaba, Niv served as an officer in the intelligence technological unit - 81, where he performed a variety of roles through the years, from hardware design, to algorithmic research and cyber security. Niv holds BSc degrees both in Electrical Engineering and in Physics (Cum Laude, 'Psagot' program) and a MSc in Optimization and Machine Learning, all from the Technion.