A Decision Tree Framework for Spatiotemporal Sequence Prediction

Taehwan Kim (Lead Author), Yisong Yue, Sarah Taylor, Iain Matthews

Research output: Chapter in Book/Report/Conference proceedingConference contribution

37 Citations (Scopus)

Abstract

We study the problem of learning to predict a spatiotemporal output sequence given an input sequence. In contrast to conventional sequence prediction problems such as part-of-speech tagging (where output sequences are selected using a relatively small set of discrete labels), our goal is to predict sequences that lie within a high-dimensional continuous output space. We present a decision tree framework for learning an accurate non-parametric spatiotemporal sequence predictor. Our approach enjoys several attractive properties, including ease of training, fast performance at test time, and the ability to robustly tolerate corrupted training data using a novel latent variable approach. We evaluate on several datasets, and demonstrate substantial improvements over existing decision tree based sequence learning frameworks such as SEARN and DAgger.
Original languageEnglish
Title of host publicationKDD '15 Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
PublisherAssociation for Computing Machinery (ACM)
Pages577-586
ISBN (Print)978-1-4503-3664-2
DOIs
Publication statusPublished - 10 Aug 2015
Event21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - Sydney, Australia
Duration: 10 Aug 201513 Aug 2015
http://www.kdd.org/kdd2015/

Conference

Conference21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
CountryAustralia
CitySydney
Period10/08/1513/08/15
Internet address

Cite this