How the Training Procedure Impacts the Performance of Deep Learning-based Vulnerability Patching.
Antonio MastropaoloVittoria NardoneGabriele BavotaMassimiliano Di PentaPublished in: EASE (2024)
Keyphrases
- deep learning
- training procedure
- machine learning
- semidefinite programming
- support vector machine
- neural network
- unsupervised learning
- principal component analysis
- learning algorithm
- expectation maximization
- mental models
- multiple kernel learning
- em algorithm
- generative model
- weakly supervised
- feature extraction
- multiscale
- object recognition
- higher order
- dimensionality reduction