Combines automata learning and model-based testing to generate training data for recurrent neural networks modeling hybrid systems, yielding fivefold lower crash-detection error on a platooning scenario with up to 1000x fewer samples than random data.
In: Machine Learning for Dynamic Software Analysis: Potentials and Limits - International Dagstuhl Seminar 16172, Dagstuhl Castle, Germany, April 24-27, 2016, Revised Papers
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Learning a Behavior Model of Hybrid Systems Through Combining Model-Based Testing and Machine Learning (Full Version)
Combines automata learning and model-based testing to generate training data for recurrent neural networks modeling hybrid systems, yielding fivefold lower crash-detection error on a platooning scenario with up to 1000x fewer samples than random data.