Simplifying Neural Networks using Formal Verification
arXiv:1910.12396
Abstract
Deep neural network (DNN) verification is an emerging field, with diverse verification engines quickly becoming available. Demonstrating the effectiveness of these engines on real-world DNNs is an important step towards their wider adoption. We present a tool that can leverage existing verification engines in performing a novel application: neural network simplification, through the reduction of the size of a DNN without harming its accuracy. We report on the work-flow of the simplification process, and demonstrate its potential significance and applicability on a family of real-world DNNs for aircraft collision avoidance, whose sizes we were able to reduce by as much as 10%.
arXiv commentThis paper appeared at NFM 2020
References in corpus (13)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
- SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
- Natural Language Processing (almost) from Scratch
- End to End Learning for Self-Driving Cars
- Evaluating Robustness of Neural Networks with Mixed Integer Programming
- Formal Security Analysis of Neural Networks using Symbolic Intervals
- An approach to reachability analysis for feed-forward ReLU neural networks
- Deep Neural Network Compression for Aircraft Collision Avoidance Systems
- Towards Proving the Adversarial Robustness of Deep Neural Networks
- Output Range Analysis for Deep Neural Networks
- DeepSafe: A Data-driven Approach for Checking Adversarial Robustness in Neural Networks
- Algorithms for Verifying Deep Neural Networks