Worst-case Satisfaction of STL Specifications Using Feedforward Neural Network Controllers: A Lagrange Multipliers Approach.
Shakiba YaghoubiGeorgios FainekosPublished in: ACM Trans. Embed. Comput. Syst. (2019)
Keyphrases
- feedforward neural networks
- lagrange multipliers
- constrained optimization
- neural network
- linear program
- back propagation
- objective function
- recurrent neural networks
- cost function
- rate distortion
- lower bound
- control system
- multilayer perceptron
- np hard
- penalty function
- linear programming
- artificial neural networks
- support vector regression
- control strategy
- hidden layer
- activation function
- reinforcement learning
- pattern recognition
- fuzzy neural network
- kernel matrix
- dynamic programming
- scalable video coding