AdversariaLib: An Open-source Library for the Security Evaluation of Machine Learning Algorithms Under Attack.
Igino CoronaBattista BiggioDavide MaiorcaPublished in: CoRR (2016)
Keyphrases
- machine learning algorithms
- open source
- benchmark data sets
- learning algorithm
- machine learning
- decision trees
- machine learning methods
- learning tasks
- predictive accuracy
- learning problems
- machine learning approaches
- random forests
- denial of service attacks
- security vulnerabilities
- information security
- machine learning systems
- cyber attacks
- security measures
- access control
- source code
- standard machine learning algorithms
- input features
- natural language processing
- security policies
- key exchange protocol
- machine learning models
- encryption algorithm
- dos attacks
- security issues
- reinforcement learning
- training data
- feature space