Ketepatan Batch Gradient Descent: Sangat stabil namun lambat. SGD: Lebih cepat tetapi rentan terhadap osilasi dan konvergensi yang tidak stabil. Mini-Batch: Lebih stabil daripada SGD dan lebih cepat dari batch penuh. 3. Kinerja pada Dataset Bes
an appropriate loss function, such as mean squared error or cross-entropy, depending on the task: ```matlab % Error calculation error = a2 - y; % difference between predicted and true output loss = sum(error.^2) / 2; % Mean squared erro
are passed through the network, layer by layer, to produce 1. an output. Backward pass: The error (difference between predicted and actual output) is 2. propagated backward through the layers, calculating gradients that are used to update the network’s weights. This iterative process adjusts the wei
