Giter VIP home page Giter VIP logo

deviation-network's Introduction

DevNet: An End-to-end Anomaly Score Learning Network

By Guansong Pang, Chunhua Shen, Anton van den Hengel. Deep anomaly detection with deviation networks (KDD19).

Brief Introduction

Deviation network (DevNet) is introduced in our KDD19 paper, which leverages a limited number of labeled anomaly data and a large set of unlabeled data to perform end-to-end anomaly score learning. It addresses a weakly supervised anomaly detection problem in that the anomalies are partially observed only and we have no labeled normal data.

Unlike other deep anomaly detection methods that focus on using data reconstruction as the driving force to learn new representations, DevNet is devised to learn the anomaly scores directly. Therefore, DevNet directly optimize the anomaly scores, whereas most of current deep anomaly detection methods optimize the feature representations. The resulting DevNet model achieves significantly better anomaly scoring than the competing deep methods. Also, due to the end-to-end anomaly scoring, DevNet can also exploit the labeled anomaly data much more effectively.

Usage

A simple example of running DevNet is shown as follows.

python devnet_kdd19.py --network_depth=2 --runs=10 --known_outliers=30 --cont_rate=0.02 --data_format=0 --output=./results.csv --dataset=`annthyroid_21feat_normalised`

See devnet_kdd19.py for more details about each argument used in this line of code.

The key packages and their versions used in our algorithm implementation are listed as follows

  • python==3.6.6
  • keras==2.2.4
  • tensorflow-gpu==1.10.0
  • scikit-learn==0.20.0
  • numpy==1.14.5
  • pandas==0.23.4
  • scipy==1.1.0
  • tensorboard==1.10.0

See the full paper below for the implemenation details of DevNet.

Full Paper

The full paper can be found at ACM Portal or arXiv

Datasets

The datasets used in DevNet are also released here. See our anomaly detection dataset repository ADRepository for more preprocessed datasets that are widely-used in other papers.

Citation

Guansong Pang, Chunhua Shen, and Anton van den Hengel. "Deep anomaly detection with deviation networks." In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp. 353-362. 2019.

deviation-network's People

Contributors

guansongpang avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.