chain rule
3 episodes mention this concept
lexfridmanSep 27, 2016Foundations of Deep Learning: Feedforward Neural Networks, Backpropagation, and Training Techniques
TechnologyFeedforward Neural NetworksInput VectorHidden LayersPre-activation- lexfridmanSep 27, 2016
Demystifying Deep Learning: A Practical and Conceptual Dive into Torch and Automatic Differentiation
TechnologyMachine LearningDeep Learning LibrariesTorch (programming framework)Lua (programming language) - lexfridman
Recurrent Neural Networks and the Fundamentals of Backpropagation for Steering Through Time
TechnologyRecurrent Neural Networks (RNNs)BackpropagationVanishing GradientsExploding Gradients
Knowledge Graph
Related concepts — line thickness indicates connection strength. Click any node to explore.
Related concepts
Concepts that appear alongside chain rule across episodes.
- stochastic gradient descent (sgd)
- backpropagation
- l2 regularization (weight decay)
- gradient-based optimization
- recurrent neural networks (rnns)
- loss function (objective function, cost function)
- flow graph
- dropout
- weights and biases
- pre-activation
- lua (programming language)
- vanishing gradients
- empirical risk minimization
- neural style transfer
- saddle points
- classification
- parameter sharing
- automatic differentiation
- learning rate
- cost function