Manifold-based Test Generation for Image Classifiers
| Autor: | Taejoon Byun, Abhishek Vijayakumar, Sanjai Rayadurgam, Darren Cofer |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
FOS: Computer and information sciences
Computer Science - Machine Learning 0209 industrial biotechnology Correctness Computer science Machine Learning (stat.ML) 02 engineering and technology Machine learning computer.software_genre Machine Learning (cs.LG) Image (mathematics) law.invention Computer Science - Software Engineering 020901 industrial engineering & automation Statistics - Machine Learning law 0202 electrical engineering electronic engineering information engineering Training set Contextual image classification Artificial neural network business.industry Autoencoder Manifold Software Engineering (cs.SE) Data point Test case 020201 artificial intelligence & image processing Artificial intelligence business Manifold (fluid mechanics) computer |
| Zdroj: | AITest |
| DOI: | 10.1109/aitest49225.2020.00010 |
| Popis: | Neural networks used for image classification tasks in critical applications must be tested with sufficient realistic data to assure their correctness. To effectively test an image classification neural network, one must obtain realistic test data adequate enough to inspire confidence that differences between the implicit requirements and the learned model would be exposed. This raises two challenges: first, an adequate subset of the data points must be carefully chosen to inspire confidence, and second, the implicit requirements must be meaningfully extrapolated to data points beyond those in the explicit training set. This paper proposes a novel framework to address these challenges. Our approach is based on the premise that patterns in a large input data space can be effectively captured in a smaller manifold space, from which similar yet novel test cases---both the input and the label---can be sampled and generated. A variant of Conditional Variational Autoencoder (CVAE) is used for capturing this manifold with a generative function, and a search technique is applied on this manifold space to efficiently find fault-revealing inputs. Experiments show that this approach enables generation of thousands of realistic yet fault-revealing test cases efficiently even for well-trained models. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|