Verification of Binarized Neural Networks via Inter-Neuron Factoring
arXiv:1710.03107
Abstract
We study the problem of formal verification of Binarized Neural Networks (BNN), which have recently been proposed as a energy-efficient alternative to traditional learning networks. The verification of BNNs, using the reduction to hardware verification, can be even more scalable by factoring computations among neurons within the same layer. By proving the NP-hardness of finding optimal factoring as well as the hardness of PTAS approximability, we design polynomial-time search heuristics to generate factoring solutions. The overall framework allows applying verification techniques to moderately-sized BNNs for embedded devices with thousands of neurons and inputs.
arXiv commentVersion 2: add proofs for hardness of PTAS approximability, remove experiments on randomized examples and some not-so-important optimizations
References in corpus (6)
- End to End Learning for Self-Driving Cars
- Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1
- An Empirical Evaluation of Deep Learning on Highway Driving
- An approach to reachability analysis for feed-forward ReLU neural networks
- Bitwise Neural Networks
- Universal adversarial perturbations
Cited by in corpus (4)
- Testing and verification of neural-network-based safety-critical control software: A systematic literature review
- Toward Scalable Verification for Safety-Critical Deep Networks
- A Statistical Approach to Assessing Neural Network Robustness
- Formal methods and software engineering for DL. Security, safety and productivity for DL systems development