Efficient top rank optimization with gradient boosting for supervised anomaly detection

Jordan Frery (1) , Amaury Habrard (1) , Marc Sebban (1) , Olivier Caelen (2) , Liyun He-Guelton (2)

1 LHC - Laboratoire Hubert Curien
2 Atos Worldline

In this paper we address the anomaly detection problem in a supervised setting where positive examples might be very sparse. We tackle this task with a learning to rank strategy by optimizing a dif-ferentiable smoothed surrogate of the so-called Average Precision (AP). Despite its non-convexity, we show how to use it efficiently in a stochas-tic gradient boosting framework. We show that using AP is much better to optimize the top rank alerts than the state of the art measures. We demonstrate on anomaly detection tasks that the interest of our method is even reinforced in highly unbalanced scenarios.

Domaines: Apprentissage [cs.LG]

Jordan Frery, Amaury Habrard, Marc Sebban, Olivier Caelen, Liyun He-Guelton. Efficient top rank optimization with gradient boosting for supervised anomaly detection. ECML PKDD 2017 European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Sep 2017, Skopje, Macedonia. ⟨hal-01611346v1⟩