On-line Training of Neural Networks: A Sliding Window Approach for the Levenberg-Marquardt Algorithm.
Fernando Morgado DiasAna AntunesJosé M. N. VieiraAlexandre Manuel MotaPublished in: IWINAC (2) (2005)
Keyphrases
- levenberg marquardt
- training algorithm
- sliding window
- neural network
- least squares
- back propagation
- artificial neural networks
- fixed size
- learning algorithm
- nonlinear least squares
- data streams
- training process
- closed form
- gauss newton
- window size
- bundle adjustment
- convergence rate
- real time
- similarity measure
- variable size
- trust region
- feature selection
- machine learning
- feed forward
- camera calibration
- semi supervised
- image features
- expert systems
- initial guess
- boyer moore
- compact data structure