A Penalty Strategy Combined Varying-Parameter Recurrent Neural Network for Solving Time-Varying Multi-Type Constrained Quadratic Programming Problems.
Zhijun ZhangSong YangLunan ZhengPublished in: IEEE Trans. Neural Networks Learn. Syst. (2021)
Keyphrases
- quadratic programming problems
- recurrent neural networks
- multi type
- equality constraints
- complex valued
- systems of linear equations
- feed forward
- data objects
- neural network
- recurrent networks
- support vector machine
- relational data
- cost model
- artificial neural networks
- penalty function
- hidden layer
- nonlinear programming
- neural nets
- database systems
- machine learning