The Forget-me-not Process
- Kieran Milan
- Joel Veness
- James Kirkpatrick
- Michael Bowling, University of Alberta
- Anna Koop, University of Alberta
- Demis Hassabis
We introduce the Forget-me-not Process, an efficient, non-parametric meta-algorithm for online probabilistic sequence prediction for piecewise stationary, repeating sources. Our method works by taking a Bayesian approach to partition a stream of data into postulated task-specific segments, while simultaneously building a model for each task. We provide regret guarantees with respect to piecewise stationary data sources under the logarithmic loss, and validate the method empirically across a range of sequence prediction and task identification problems.
Citation
K. Milan, J. Veness, J. Kirkpatrick, M. Bowling, A. Koop, D. Hassabis. "The Forget-me-not Process". Neural Information Processing Systems, (ed: Daniel D. Lee, Masashi Sugiyama, Ulrike von Luxburg, Isabelle Guyon, Roman Garnett), pp 3702-3710, December 2016.Keywords: | |
Category: | In Conference |
Web Links: | NeurIPS |
BibTeX
@incollection{Milan+al:NeurIPS16, author = {Kieran Milan and Joel Veness and James Kirkpatrick and Michael Bowling and Anna Koop and Demis Hassabis}, title = {The Forget-me-not Process}, Editor = {Daniel D. Lee, Masashi Sugiyama, Ulrike von Luxburg, Isabelle Guyon, Roman Garnett}, Pages = {3702-3710}, booktitle = {Neural Information Processing Systems}, year = 2016, }Last Updated: October 28, 2020
Submitted by Sabina P