Why do we update each weights and bias to minimize log loss in neural networks?

1 Like

We update the weights and bias values in order to reach the minimum cost.

2 Likes

means that each layer in the neural network runs multiple times in order to find the best weight value from one perceptron to another. Am I right?

please correct me. Thanks

1 Like

Yes.

1 Like

Sorry, please excuse and ignore these two messages. I posted them on the wrong thread.

My apologies.