University of Cambridge > Talks.cam > Machine Learning @ CUED > Stochastic Outlier Selection

  Browse A-Z

Log in

Cambridge users (raven)details Other usersdetails No account?details

Information on

Subscribing to talksdetails Finding a talkdetails Adding a talkdetails Disseminating talksdetails Help and Documentationdetails

Stochastic Outlier Selection

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Zoubin Ghahramani.

Anomaly detection is relevant for many tasks, ranging from detecting credit card fraud to terrorist threats. Typically, outlier-selection algorithms are used for detecting anomalies. First, we briefly explain the difference between anomalies and outliers. Subsequently, we present a novel, unsupervised outlier-selection algorithm, called Stochastic Outlier Selection (SOS). The SOS algorithm computes for each data point an outlier probability. These probabilities are much more intuitive than the unbounded outlier scores computed by existing outlier-selection algorithms. We evaluate SOS on a variety of real-world and synthetic datasets, and compare it to four state-of-the-art outlier-selection algorithms. Our results show that SOS has a superior performance while being more robust to data perturbations and parameter settings. We conclude that SOS is an effective algorithm to select outliers in a dataset that compares favorably to state-of-the-art outlier-selection algorithms.

This is work with Eric O. Postma.

http://www.jeroenjanssens.com

This talk is part of the Machine Learning @ CUED series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

Other lists

Cambridge Cleantech Cambridge Psychometrics Centre Seminars Information Structure

Other talks

Direct measurements of dynamic granular compaction at the mesoscale using synchrotron X-ray radiography Cambridge - Corporate Finance Theory Symposium September 2018 - Day 2 Chemical convection and stratification at the top of the Earth's outer core Beacon Salon #7 Imaging Far and Wide Lipschitz Global Optimization Questions of Morality in Global Health- An interdisciplinary conference Singularities of Hermitian-Yang-Mills connections and the Harder-Narasimhan-Seshadri filtration Cyclic Peptides: Building Blocks for Supramolecular Designs 'Cryptocurrency and BLOCKCHAIN – PAST, PRESENT AND FUTURE' Mathematical applications of little string theory Asclepiadaceae 'Alas, poor Yorick!': Laurence Sterne's "A Sentimental Journey" after 250 years'
 

© 2006-2024 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity