Question Bank
3705 approved questions from the community
What is a cost function in neural networks?
What is the significance of Residual Networks (ResNets)?
Why do we use convolutions for images rather than fully connected (FC) layers?
Why do segmentation CNNs typically have an encoder-decoder structure?
What is the learning rate in neural network optimization?
What is overfitting in machine learning?
What is gradient descent?
What is Principal Component Analysis (PCA)?
What is data normalization in machine learning?
What is momentum in neural network optimization?
What is the bias-variance tradeoff in machine learning?
What is the primary advantage of using Adam optimizer over SGD?
In a GAN, what happens during training?
Increasing batch size always leads to faster training and better model performance.
Learning rate should remain constant throughout training for optimal convergence.
What is the Turing Test?
What is the dying ReLU problem?
What is Non-Maximum Suppression (NMS) and how can it be optimized?
In ensemble methods, using identical models trained on different data subsets always produces better results than using diverse model architectures.
What is curriculum learning in machine learning?