Learning a Neural-network-based Representation for Open Set Recognition (2018-02-12T00:00:00.000000Z)

TL;DR

In this representation instances from the same class are close to each other while instances from different classes are further apart, resulting in statistically significant improvement when compared to other approaches on three datasets from two different domains.

Abstract

Open set recognition problems exist in many domains. For example in security, new malware classes emerge regularly; therefore malware classification systems need to identify instances from unknown classes in addition to discriminating between known classes. In this paper we present a neural network based representation for addressing the open set recognition problem. In this representation instances from the same class are close to each other while instances from different classes are further apart, resulting in statistically significant improvement when compared to other approaches on three datasets from two different domains.

Authors

Mehadi Hassen

1 papers

P. Chan

1 papers

Learning a Neural-network-based Representation for Open Set Recognition

TL;DR

Abstract

Authors

References25 items

Generative Adversarial Networks

Microsoft Malware Classification Challenge

Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samples

Steps Toward Robust Artificial Intelligence

Generative OpenMax for Multi-Class Open Set Classification

Open Category Classification by Adversarial Sample Generation

Scalable Function Call Graph-based Malware Classification

A Survey of Stealth Malware Attacks, Mitigation Measures, and Steps Toward Autonomous Open World Solutions

Towards Open Set Deep Networks

Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks

Adversarial Autoencoders

A bounded neural network for open set recognition

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Local Novelty Detection in Multi-class Recognition Problems

Towards Open World Recognition

Multi-class Open Set Recognition Using Probability of Inclusion

Learning with Augmented Class by Exploiting Unlabeled Data

Toward Open Set Recognition

Kernel Null Space Methods for Novelty Detection

Automatic analysis of malware behavior using machine learning

Adagio

Unlabeled Data

Face recognition for web-scale datasets

Mnist hand wri en digit dataset

Android malware genome project

Field of Study

Venue Information

Name

Type

URL

Alternate Names