What is ‘gradient descent’ in neural network training?

Question

AI Fundamentals — Medium

What is ‘gradient descent’ in neural network training?

Accepted Answer

Gradient descent in neural network training is an optimization algorithm that adjusts model parameters to minimize the loss function. By iteratively updating the parameters in the direction of steepest descent, the algorithm helps the model converge towards the optimal solution.