Automated Verification of Neural Networks: Advances, Challenges and Perspectives
arXiv:1805.09938
Abstract
Neural networks are one of the most investigated and widely used techniques in Machine Learning. In spite of their success, they still find limited application in safety- and security-related contexts, wherein assurance about networks' performances must be provided. In the recent past, automated reasoning techniques have been proposed by several researchers to close the gap between neural networks and applications requiring formal guarantees about their behavior. In this work, we propose a primer of such techniques and a comprehensive categorization of existing approaches for the automated verification of neural networks. A discussion about current limitations and directions for future investigation is provided to foster research on this topic at the crossroads of Machine Learning and Automated Reasoning.
References in corpus (6)
- End to End Learning for Self-Driving Cars
- BinaryConnect: Training Deep Neural Networks with binary weights during propagations
- Adversarial Examples Are Not Easily Detected: Bypassing Ten Detection Methods
- Deep Neural Networks as 0-1 Mixed Integer Linear Programs: A Feasibility Study
- DeepSafe: A Data-driven Approach for Checking Adversarial Robustness in Neural Networks
- Toward Scalable Verification for Safety-Critical Deep Networks
Cited by in corpus (10)
- Optimization Problems for Machine Learning: A Survey
- Algorithms for Verifying Deep Neural Networks
- On Symbolically Encoding the Behavior of Random Forests
- Verification of Recurrent Neural Networks Through Rule Extraction
- On Tractable Representations of Binary Neural Networks
- Verification of Neural Networks: Enhancing Scalability through Pruning
- From Shallow to Deep Interactions Between Knowledge Representation, Reasoning and Machine Learning (Kay R. Amel group)
- The Power of Typed Affine Decision Structures: A Case Study
- Formal methods and software engineering for DL. Security, safety and productivity for DL systems development
- NeVer 2.0: Learning, Verification and Repair of Deep Neural Networks