|
Size: 3640
Comment:
|
Size: 4696
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 22: | Line 22: |
| Jeder Vortrag soll 35 Minuten (+ 10 Minuten Diskussion) dauern. Ein guter Vortrag führt in das jeweilige Thema ein und stellt relevante Arbeiten zusammenfassend dar. | Jeder Vortrag soll 35 Minuten (+ 10 Minuten Diskussion) dauern. Ein guter Vortrag führt kurz in das jeweilige Thema ein, stellt die Problemstellung dar und beschreibt zusammenfassend relevante Arbeiten und Lösungen. |
| Line 32: | Line 32: |
| * Anomaly Detection of Web-based Attacks. Kruegel, Vigna. ''Proceedings of CCS'' 2003 | * [[http://www.cs.ucsb.edu/~vigna/publications/2003_kruegel_vigna_ccs03.pdf|Anomaly Detection of Web-based Attacks]].<<BR>> Kruegel, Vigna. ''Proceedings of CCS'' 2003 |
| Line 34: | Line 34: |
| * Spectrogram: A Mixture-of-Markov-Chains Model for Anomaly Detection in Web Traffic. Song, Keromytis, Stolfo. ''Proceedings of NDSS'' 2009 | * [[http://www.isoc.org/isoc/conferences/ndss/09/pdf/07.pdf|Spectrogram: A Mixture-of-Markov-Chains Model for Anomaly Detection in Web Traffic]].<<BR>> Song, Keromytis, Stolfo. ''Proceedings of NDSS'' 2009 |
| Line 36: | Line 36: |
| * '''Network Intrusion Detection''' | * '''Detection of Attacks in Network Content''' |
| Line 38: | Line 38: |
| * Detecting Unknown Network Attacks Using Language Models. Rieck, Laskov. ''Proceedings of DIMVA'' 2006 | * [[http://user.cs.tu-berlin.de/~rieck/docs/2006-dimva.pdf|Detecting Unknown Network Attacks Using Language Models]].<<BR>> Rieck, Laskov. ''Proceedings of DIMVA'' 2006 |
| Line 40: | Line 40: |
| * McPAD: A Multiple Classifier System for Accurate Payload-Based Anomaly Detection. Perdisci, Ariu, Fogla, Giacinto, Lee. ''Computer Networks'' 2009 | * [[http://www.scs.carleton.ca/~soma/id-2007w/readings|Anagram: A Content Anomaly Detector Resistant to Mimicry Attack/anagram-raid2006.pd|]]. <<BR>> Wang, Parekh, Stolfo. ''Proceedings of RAID'' 2006 |
| Line 44: | Line 44: |
| * Scalable, Behavior-Based Malware Clustering. Bayer, Comparetti, Hlauscheck, Kruegel, Krida. ''Proceedings of NDSS'' 2009 | * [[http://www.iseclab.org/papers/clustering_ndss.pdf|Scalable, Behavior-Based Malware Clustering]].<<BR>> Bayer, Comparetti, Hlauscheck, Kruegel, Krida. ''Proceedings of NDSS'' 2009 |
| Line 46: | Line 46: |
| * Learning and Classification of Malware Behavior. Rieck, Holz, Willems, Düssel, Laskov. ''Proceedings of DIMVA'' 2008 | * [[http://user.cs.tu-berlin.de/~rieck/docs/2008-dimva.pdf|Learning and Classification of Malware Behavior]].<<BR>> Rieck, Holz, Willems, Düssel, Laskov. ''Proceedings of DIMVA'' 2008 |
| Line 50: | Line 50: |
| * Learning to Detect and Classify Malicious Executables in the Wild. Kolter and Maloof. ''Journal of Machine Learning Research'' 2006. | * [[http://jmlr.csail.mit.edu/papers/volume7/kolter06a/kolter06a.pdf|Learning to Detect and Classify Malicious Executables in the Wild]].<<BR>> Kolter and Maloof. ''Journal of Machine Learning Research'' 2006. |
| Line 52: | Line 52: |
| * McBoost: Boosting Scalability in Malware Collection and Analysis Using Statistical Classification of Executables. Perdisci, Lanzi, Lee. ''Proceeding of ACSAC'' 2008 | * [[http://roberto.perdisci.googlepages.com/Perdisci-McBoost.pdf|McBoost: Boosting Scalability in Malware Collection and Analysis Using Statistical Classification of Executables]].<<BR>> Perdisci, Lanzi, Lee. ''Proceeding of ACSAC'' 2008 |
| Line 54: | Line 54: |
| * '''Mimicry Attacks''' | * '''Mimicry and Evasion Attacks''' |
| Line 56: | Line 56: |
| * Polymorphic Blending Attacks. Fogla, Sharif, Perdisci, Kolesnikov, Lee. ''Proceedings of USENIX SS'' 2006 | * [[http://www.cc.gatech.edu/~wenke/papers/usenix_security_2006.pdf|Polymorphic Blending Attacks]].<<BR>> Fogla, Sharif, Perdisci, Kolesnikov, Lee. ''Proceedings of USENIX SS'' 2006 |
| Line 58: | Line 58: |
| * English Shellcode. Mason, Small, Monrose, Macmanus. ''Proceedings of CCS'' 2009 | * [[http://www.cs.jhu.edu/~sam/ccs243-mason.pdf|English Shellcode]].<<BR>> Mason, Small, Monrose, Macmanus. ''Proceedings of CCS'' 2009 |
| Line 62: | Line 62: |
| * Polygraph: Automatically Generating Signatures for Polymorphic Worms. Newsome, Karp, Song. ''Proceedings of IEEE S&P'' 2005 | * [[http://jimnewsome.net/papers/polygraph-oakland2005.pdf|Polygraph: Automatically Generating Signatures for Polymorphic Worms]].<<BR>> Newsome, Karp, Song. ''Proceedings of IEEE S&P'' 2005 |
| Line 64: | Line 64: |
| * Hamsa: Fast Signature Generation for Zero-day Polymorphic Worms with Provable Attack Resilience. Li, Sanghi, Chen, Kao, Chavez. ''Proceedings of IEEE S&P'' 2006 | * [[http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.91.7756&rep=rep1&type=pdf|Hamsa: Fast Signature Generation for Zero-day Polymorphic Worms with Provable Attack Resilience]].<<BR>> Li, Sanghi, Chen, Kao, Chavez. ''Proceedings of IEEE S&P'' 2006 |
| Line 68: | Line 68: |
| * Limits of Learning-based Signature Generation with Adversaries. Blum, Song. ''Proceedings of NDSS'' 2008 | * [[http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.122.9397&rep=rep1&type=pdf|Limits of Learning-based Signature Generation with Adversaries]].<<BR>> Blum, Song. ''Proceedings of NDSS'' 2008 |
| Line 70: | Line 70: |
| * Misleading Worm Signature Generators Using Deliberate Noise Injection. Perdicsi, Dagon, Lee, Fogla, Sharif. ''Proceedings of IEEE S&P'' 2006 | * [[http://www.cc.gatech.edu/~wenke/papers/ieee-sp-06.pdf|Misleading Worm Signature Generators Using Deliberate Noise Injection]].<<BR>> Perdicsi, Dagon, Lee, Fogla, Sharif. ''Proceedings of IEEE S&P'' 2006 |
Seminar: Machinelles Lernen in der IT-Sicherheit
Termine und Dozenten
Termin |
Vorbesprechung und Themenvergabe am X.X.2010 um 14:00 Uhr im Raum FR 6046 |
|
Blockseminar am X.X.2010 von 10:00 bis 16:00 Uhr im Raum FR 6046 |
Verantwortlich |
Inhalt
In diesem Seminar beschäftigen wir uns mit Anwendungen des maschinellen Lernens in der IT-Sicherheit.
Ablauf
- Die Vorbesprechung findet am X.X.2010 statt.
- Die Teilnehmer wählen bis spätestens X.X.2010 ein Thema in Absprache mit dem Betreuer (siehe Themenliste).
- Die Teilnehmer legen bis spätestens X.X.2010 einen Entwurf des Vortrages vor und besprechen diesen mit dem Betreuer.
- Das Seminar findet als Blockveranstaltung am X.X.2.2010 statt.
Vorträge
Jeder Vortrag soll 35 Minuten (+ 10 Minuten Diskussion) dauern. Ein guter Vortrag führt kurz in das jeweilige Thema ein, stellt die Problemstellung dar und beschreibt zusammenfassend relevante Arbeiten und Lösungen.
Leistungsnachweis
Die Note wird anhand des Vortrages und der Folien festgelegt. Das Seminar ist Wahlpflichtbestandteil der Master-Module "Maschinelles Lernen 1" und "Maschinelles Lernen 2". Bachelor-Studenten können diese Master-Module auf Antrag ebenfalls belegen. Wir empfehlen den vorherigen Besuch der Vorlesung "Maschinelles Lernen 1".
Themen
Detection of Web-based Attacks
Anomaly Detection of Web-based Attacks.
Kruegel, Vigna. Proceedings of CCS 2003Spectrogram: A Mixture-of-Markov-Chains Model for Anomaly Detection in Web Traffic.
Song, Keromytis, Stolfo. Proceedings of NDSS 2009
Detection of Attacks in Network Content
Detecting Unknown Network Attacks Using Language Models.
Rieck, Laskov. Proceedings of DIMVA 2006Anagram: A Content Anomaly Detector Resistant to Mimicry Attack/anagram-raid2006.pd.
Wang, Parekh, Stolfo. Proceedings of RAID 2006
Analysis of Malicious Program Behavior
Scalable, Behavior-Based Malware Clustering.
Bayer, Comparetti, Hlauscheck, Kruegel, Krida. Proceedings of NDSS 2009Learning and Classification of Malware Behavior.
Rieck, Holz, Willems, Düssel, Laskov. Proceedings of DIMVA 2008
Analysis of Malicious Executables
Learning to Detect and Classify Malicious Executables in the Wild.
Kolter and Maloof. Journal of Machine Learning Research 2006.McBoost: Boosting Scalability in Malware Collection and Analysis Using Statistical Classification of Executables.
Perdisci, Lanzi, Lee. Proceeding of ACSAC 2008
Mimicry and Evasion Attacks
Polymorphic Blending Attacks.
Fogla, Sharif, Perdisci, Kolesnikov, Lee. Proceedings of USENIX SS 2006English Shellcode.
Mason, Small, Monrose, Macmanus. Proceedings of CCS 2009
Automatic Signature Generation
Polygraph: Automatically Generating Signatures for Polymorphic Worms.
Newsome, Karp, Song. Proceedings of IEEE S&P 2005Hamsa: Fast Signature Generation for Zero-day Polymorphic Worms with Provable Attack Resilience.
Li, Sanghi, Chen, Kao, Chavez. Proceedings of IEEE S&P 2006
Attacks against Signature Generation
Limits of Learning-based Signature Generation with Adversaries.
Blum, Song. Proceedings of NDSS 2008Misleading Worm Signature Generators Using Deliberate Noise Injection.
Perdicsi, Dagon, Lee, Fogla, Sharif. Proceedings of IEEE S&P 2006