`

Timezone: »

 
Poster
A loss framework for calibrated anomaly detection
Aditya Menon · Robert Williamson

Wed Dec 05 07:45 AM -- 09:45 AM (PST) @ Room 517 AB #110
in Wed Poster Session A »

Given samples from a probability distribution, anomaly detection is the problem of determining if a given point lies in a low-density region. This paper concerns calibrated anomaly detection, which is the practically relevant extension where we additionally wish to produce a confidence score for a point being anomalous. Building on a classification framework for anomaly detection, we show how minimisation of a suitably modified proper loss produces density estimates only for anomalous instances. We then show how to incorporate quantile control by relating our objective to a generalised version of the pinball loss. Finally, we show how to efficiently optimise the objective with kernelised scorer, by leveraging a recent result from the point process literature. The resulting objective captures a close relative of the one-class SVM as a special case.

Keywords: Unsupervised Learning · Learning Theory
[ Paper] 

Author Information

Aditya Menon (Google Research)
Robert Williamson (Australian National University & Data61)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors