Sleeper Agent: Scalable Hidden Trigger Backdoors for Neural Networks Trained from Scratch.
Hossein SouriMicah GoldblumLiam FowlRama ChellappaTom GoldsteinPublished in: CoRR (2021)
Keyphrases
- neural network
- multilayer perceptron
- trained neural networks
- training process
- multi layer perceptron
- auto associative
- multiagent systems
- multi agent systems
- trained neural network
- pattern recognition
- multi agent
- backpropagation algorithm
- artificial neural networks
- autonomous agents
- intelligent agents
- software agents
- back propagation
- agent systems
- rule extraction
- neural learning
- agent technology
- mobile agents
- fuzzy logic
- genetic algorithm
- hidden units
- multilayer neural network
- learning capabilities
- pedagogical agents
- decision making
- agent architecture
- feedforward neural networks
- satisfiability problem
- recurrent neural networks
- feed forward
- self organizing maps
- dynamic environments
- neural network is trained
- neural nets
- cognitive agents
- artificial agents
- neural network model
- training data