Temporal SegmentatioN:

perspectives from statistics, machine learning, and signal processing.

December 12th 2009, Whistler, British Columbia, CA

Sponsored by Pascal II

Organizing Committee

Workshop Overview

Invited Speakers

Call for papers



Organizing Committee

Workshop Overview

Data with temporal (or sequential) structure arise in several applications, such as speaker diarization [FSJW08b, FDH08], human action segmentation [ZTH08], network intrusion detection [TRBK06], DNA copy number analysis [LXZ08], and neuron activity modelling [Y07], to name a few.

A particularly recurrent temporal structure in real applications is the so-called change-point model [BH92], where the data may be temporally partitioned into a sequence of segments delimited by change-points, such that a single model holds within each segment whereas different models hold accross segments. Change-point problems may be tackled from two points of view, corresponding to the practical problem at hand: retrospective (or "a posteriori"), aka multiple change-point estimation [F06], where the whole signal is taken at once and the goal is to estimate the change-point locations [BKLMW09], and online (or sequential), aka quickest detection [PH09], where data are observed sequentially and the goal is to quickly detect change-points. We refer to these classes of tasks as temporal segmentation.

An extensive literature has developed in these two viewpoints, in both the statistics (and probability) community [L01], and in the signal processing community [K98]. Many of the optimal algorithms proposed in this literature were developed under rather restrictive assumptions, however: parametric models for distributions, low-dimensional multivariate data, and, in the online case, perfect knowledge of the pre- and post-change distributions.

In applications such as human action segmentation or speaker diarization, data are large-scale, expensive to label, and high-dimensional, therefore requiring approaches that can tackle more complex situations in temporal segmentation. Recent years have witnessed new approaches with broader applicability, essentially by proposing unsupervised [XWSS06, ZTH08], nonparametric [FSJW08b], and scalable temporal segmentation algorithms [FL07, FDH08].

The purpose of this workshop is to bring together experts from the statistics, machine learning, signal processing communities, to address a broad range of applications from robotics to neuroscience, to discuss and cross-fertilize ideas, and to define the current challenges in temporal segmentation. We intend to encourage discussions on the following particular issues: How can traditional statistical approaches for temporal segmentation, essentially generative, be extended to discriminative approaches, allowing us to deal with high-dimensional data? How well do unsupervised approaches for temporal segmentation perform with respect to supervised ones? What are the main statistical and computational issues that arise when addressing large-scale (long) data signals?

Invited speakers

Call for papers

Submissions are solicited for the workshop. Submissions are not constrained regarding the applications in which temporal segmentation is of interest: these can include speaker/audio processing, financial time series, video, motion capture, bioinformatics, network intrusion detection, etc. Presentations on closely related topics such as sequential decision making, temporal classification, structured prediction with temporal/spatial constraints, are also encouraged. Submissions may focus on theoretical, methodological, and computational aspects of temporal segmentation.

Accepted submissions will be presented either as short talks or during the workshop poster session, depending on the proposals. The deadline for submission is October 23th 2009 (Midnight, Pacific Standard Time) and notifications will be sent out by November 2nd 2009. The submission should be at most four pages long in NIPS format, and should be sent to temposegment.nips09@gmail.com. We do not require submissions to be in blind format. Yet, submissions may be in blind format if the authors wish to do so.

Any question regarding the workshop and the submission/reviewing process may be sent to zaidh@andrew.cmu.edu.






Haipeng Xing &Tze Leung Lai

Retrospective and Sequential Change-point Modelling Approaches


Jens Kohlmorgen

A Dynamic HMM for Online Segmentation


Discussions and posters


Kevin Murphy

Product Partition Models for Modelling Changing Dependency Structure in Time Series 


Fernando de la Torre

Segmentation of Human Activities in Video


Erik Sudderth

Hierarchical-Dirichlet-Process-based Hidden Markov Models


Posters and discussion


Olympia Hadjiliadis

Sequential Change-point Detection


Alexandre Lung-Yut-Fong

Distributed Detection and Localization of Network Anomalies using Rank Tests


Brian Williams

Mode estimation of Autonomous Systems


Discussion and posters


Ryan Turner

Adaptive Sequential Bayesian Change-point Detection


Francis Bach

Temporal Segmentation with Kernel Change-point Detection


Discussion and wrap-up