Knowledge Graph — Coursera Notes › Academic disciplines › Computer Science / Information Technology › Artificial Intelligence › Neural Network

Backpropagation

concept · part of Neural Network

An algorithm that adjusts weights in a neural network based on the error computed during forward propagation to minimize loss. Backpropagation is the training method for feedforward networks where error is propagated backward to update weights, making it essential for learning in neural networks.

Inside Backpropagation (1)

• Gradient descent — An optimization algorithm used in backpropagation to adjust weights in the direction that minimizes the loss.

Connections

• Used for Neural Network
• Uses Gradient descent

This is the text view of an interactive 3D knowledge graph — open this page with JavaScript enabled to explore it visually.