Titolo:  One-and-a-half-class multiple classifier systems for secure learning against evasion attacks at test time
Autori: 
Data di pubblicazione:  2015
Autori:  Biggio, Battista; Corona, Igino; He, Z. M.; Chan P., P; Giacinto, Giorgio; Yeung D., S; Roli, Fabio
Presenza coautori internazionali: 
Lingua:  Inglese
Titolo del libro:  Multiple Classifier Systems
ISBN:  978-3-319-20247-1
978-3-319-20248-8
978-3-319-20247-1
978-3-319-20248-8
Editore:  Springer Verlag
Tutti i curatori:  Friedhelm Schwenker, Fabio Roli, Josef Kittler
Serie:  LECTURE NOTES IN COMPUTER SCIENCE
Volume:  9132
Pagina iniziale:  168
Pagina finale:  180
Numero di pagine:  13
Digital Object Identifier (DOI):  http://dx.doi.org/10.1007/978-3-319-20248-8_15
Codice identificativo Scopus:  2-s2.0-84937460019
Codice identificativo ISI:  WOS:000364539000015
Revisione (peer review):  Esperti anonimi
Nome del convegno:  12th International Workshop, MCS 2015
Periodo del convegno:  June 29 - July 1, 2015
Luogo del convegno:  Günzburg, Germany,
Abstract:  Pattern classifiers have been widely used in adversarial settings like spam and malware detection, although they have not been originally designed to cope with intelligent attackers that manipulate data at test time to evade detection. While a number of adversary-aware learning algorithms have been proposed, they are computationally demanding and aim to counter specific kinds of adversarial data manipulation. In this work, we overcome these limitations by proposing a multiple classifier system capable of improving security against evasion attacks at test time by learning a decision function that more tightly encloses the legitimate samples in feature space, without significantly compromising accuracy in the absence of attack. Since we combine a set of one-class and two-class classifiers to this end, we name our approach one-and-a-halfclass (1.5C) classification. Our proposal is general and it can be used to improve the security of any classifier against evasion attacks at test time, as shown by the reported experiments on spam and malware detection
Tipologia: 4.1 Contributo in Atti di convegno

File in questo prodotto:
File Descrizione Tipologia Licenza  
chp%3A10.1007%2F978-3-319-20248-8_15.pdf  versione editoriale Administrator   Richiedi una copia

Questionario e social

Condividi su: