The Marabou Framework for Verification and Analysis of Deep Neural Networks

“The Marabou Framework for Verification and Analysis of Deep Neural Networks” by Guy Katz, Derek A. Huang, Duligur Ibeling, Kyle Julian, Christopher Lazarus, Rachel Lim, Parth Shah, Shantanu Thakoor, Haoze Wu, Aleksandar Zeljic, David L. Dill, Mykel J. Kochenderfer, and Clark Barrett. In Proceedings of the 31^st International Conference on Computer Aided Verification (CAV '19), (Isil Dillig and Serdar Tasiran, eds.), July 2019, pp. 443-452. New York, New York.

Abstract

Deep neural networks are revolutionizing the way complex systems are designed. Consequently, there is a pressing need for tools and techniques for network analysis and certification. To help in addressing that need, we present Marabou, a framework for verifying deep neural networks. Marabou is an SMT-based tool that can answer queries about a network's properties by transforming these queries into constraint satisfaction problems. It can accommodate networks with different activation functions and topologies, and it performs high-level reasoning on the network that can curtail the search space and improve performance. It also supports parallel execution to further enhance scalability. Marabou accepts multiple input formats, including protocol buffer files generated by the popular TensorFlow framework for neural networks. We describe the system architecture and main components, evaluate the technique and discuss ongoing work.

BibTeX entry:

@inproceedings{KHI+19,
   author = {Guy Katz and Derek A. Huang and Duligur Ibeling and Kyle
	Julian and Christopher Lazarus and Rachel Lim and Parth Shah and
	Shantanu Thakoor and Haoze Wu and Aleksandar Zelji{\'c} and David
	L. Dill and Mykel J. Kochenderfer and Clark Barrett},
   editor = {Isil Dillig and Serdar Tasiran},
   title = {The Marabou Framework for Verification and Analysis of Deep
	Neural Networks},
   booktitle = {Proceedings of the {\it 31^{st}} International Conference
	on Computer Aided Verification (CAV '19)},
   series = {Lecture Notes in Computer Science},
   volume = {11561},
   pages = {443--452},
   publisher = {Springer International Publishing},
   month = jul,
   year = {2019},
   isbn = {978-3-030-25539-8},
   doi = {10.1007/978-3-030-25540-4_26},
   note = {New York, New York},
   url = {http://theory.stanford.edu/~barrett/pubs/KHI+19.pdf}
}

(This webpage was created with bibtex2web.)