Poster (Scientific congresses, symposiums and conference proceedings)
Human in the Loop: Interactive Passive Automata Learning via Evidence-Driven State-Merging Algorithms
Hammerschmidt, Christian; State, Radu; Verwer, Sicco
2017Human in the Loop Machine Learning Workshop at the International Conference on Machine Learning
 

Files


Full Text
untitled (4).pdf
Publisher postprint (195.51 kB)
Download

All documents in ORBilu are protected by a user license.

Send to



Details



Keywords :
Computer Science - Learning; Statistics - Machine Learning
Abstract :
[en] We present an interactive version of an evidence-driven state-merging (EDSM) algorithm for learning variants of finite state automata. Learning these automata often amounts to recovering or reverse engineering the model generating the data despite noisy, incomplete, or imperfectly sampled data sources rather than optimizing a purely numeric target function. Domain expertise and human knowledge about the target domain can guide this process, and typically is captured in parameter settings. Often, domain expertise is subconscious and not expressed explicitly. Directly interacting with the learning algorithm makes it easier to utilize this knowledge effectively.
Disciplines :
Computer science
Author, co-author :
Hammerschmidt, Christian ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT)
State, Radu  ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT)
Verwer, Sicco;  Delft University of Technology - TU Delft > Cyber Security Group
External co-authors :
no
Language :
English
Title :
Human in the Loop: Interactive Passive Automata Learning via Evidence-Driven State-Merging Algorithms
Publication date :
August 2017
Event name :
Human in the Loop Machine Learning Workshop at the International Conference on Machine Learning
Event place :
Sydney, Australia
Event date :
August 11th
Audience :
International
Focus Area :
Computational Sciences
FnR Project :
FNR10053360 - Stream Mining For Predictive Authentication Under Adversarial Influence, 2015 (01/03/2015-14/11/2017) - Christian Hammerschmidt
Commentary :
arXiv: 1707.09430
Available on ORBilu :
since 05 November 2017

Statistics


Number of views
74 (1 by Unilu)
Number of downloads
65 (1 by Unilu)

Bibliography


Similar publications



Contact ORBilu