ss2163.md

title	date	firstname	lastname	email	year_arrived	year_left	still_around	status	website	google_scholar	github	twitter	linkedin	labs
Siddharth Swaroop	2020-12-14 09:32:38 UTC	Siddharth	Swaroop	[email protected]	2017		true	student	https://siddharthswaroop.github.io/	https://scholar.google.com/citations?user=zl8B-9oAAAAJ	siddharthswaroop	siddharthswar	siddharthswaroop	cbl turner

Siddharth is a PhD candidate supervised by Prof Richard E Turner and advised by Prof Carl Rasmussen. He joined the group in 2017, after receiving an MEng from University of Cambridge. His PhD is funded by the EPSRC, as well as a Microsoft Research EMEA PhD Award.

He is interested in designing algorithms for large-scale machine learning systems that learn sequentially without revisiting past data, are private over user data, and are uncertainty-aware. He uses probabilistic methods, and has focussed on approximate Bayesian inference techniques, usually variational inference. His research scales these techniques to large Bayesian neural network models, and applies these to the continual learning (/lifelong learning) and federated learning problems.