Strengthening Gradient Descent by Sequential Motion Optimization for Deep Neural Networks.
Thang Le-DucQuoc-Hung NguyenJaehong LeeHung Nguyen-XuanPublished in: IEEE Trans. Evol. Comput. (2023)
Keyphrases
- neural network
- pattern recognition
- motion estimation
- optimization algorithm
- motion analysis
- image sequences
- optimization problems
- motion model
- cost function
- human motion
- learning rules
- motion capture
- stochastic gradient descent
- optimization process
- motion planning
- global optimization
- spatial and temporal
- space time
- back propagation
- motion detection
- multi view
- fuzzy logic
- moving objects
- camera motion
- neural network model
- feed forward
- motion field
- motion segmentation
- humanoid robot
- constrained optimization
- artificial neural networks
- support vector