Deakin University
Browse

File(s) under permanent embargo

Regret bounds for transfer learning in Bayesian optimisation

conference contribution
posted on 2017-04-29, 00:00 authored by Alistair ShiltonAlistair Shilton, Sunil GuptaSunil Gupta, Santu RanaSantu Rana, Svetha VenkateshSvetha Venkatesh
This paper studies the regret bound of two transfer learning algorithms in Bayesian optimisation. The first algorithm models any difference between the source and target functions as a noise process. The second algorithm proposes a new way to model the difference between the source and target as a Gaussian process which is then used to adapt the source data. We show that in both cases the regret bounds are tighter than in the no transfer case. We also experimentally compare the performance of these algorithms relative to no transfer learning and demonstrate benefits of transfer learning.

History

Volume

54

Pagination

1-9

Location

Fort Lauderdale, Florida

Start date

2017-04-20

End date

2017-04-22

Language

eng

Publication classification

E Conference publication, E1 Full written paper - refereed

Copyright notice

2017, The Authors

Editor/Contributor(s)

Singh A, Zhu J

Title of proceedings

AISTATS 2017 : Machine Learning Research : Proceedings of the 20th Artificial Intelligence and Statistics International Conference

Event

Artificial Intelligence and Statistics. International Conference (20th : 2017 : Fort Lauderdale, Florida)

Publisher

[The Conference]

Place of publication

[Fort Lauderdale, Fla.]