NCA-GENM - Core Machine Learning and AI Knowledge - Section 2.1

Understand fundamental techniques for training deep learning models.

Define the role of loss functions, backpropagation, and gradient descent in iteratively updating neural network weights. Recognise common training pitfalls such as overfitting, vanishing gradients, and the use of regularisation to address them.

Practice question for this objective

Free sampleCore Machine Learning and AI Knowledgemedium

A training run shows that validation loss stops improving after epoch 10 while training loss continues to fall steeply. Which technique directly addresses this gap by penalising large weight magnitudes in the loss function itself?

  • AInserting batch normalisation layers after each linear block, which normalises activations across the mini-batch during the forward pass.
  • BApplying L2 weight decay, which adds the sum of squared weights to the training loss, discouraging large parameter values. Correct
  • CSwitching from a cosine annealing schedule to a step decay schedule so the learning rate drops more sharply every few epochs.
  • DReplacing stochastic gradient descent with Adam, which adapts per-parameter learning rates using first and second moment estimates.
Identify which regularisation technique directly modifies the loss function to discourage overfitting in a deep learning model. L2 weight decay appends a term equal to the learning rate multiplied by the sum of squared weights to the gradient update, effectively shrinking weights each step. This penalises model complexity directly in the loss, reducing overfitting when training loss diverges from validation loss.

Why A is wrong: Batch normalisation stabilises training and can reduce internal covariate shift, but it does not add a penalty on weight magnitudes to the loss function itself, so it addresses a different cause of the training-validation gap.

Why B is correct: L2 regularisation adds a penalty proportional to the squared magnitude of weights directly to the loss, reducing overfitting by keeping weights small and the model less sensitive to training-set noise.

Why C is wrong: Learning-rate schedules control how step size evolves during optimisation, which can help convergence but does not add a regularisation penalty on weights to the loss and will not directly close the gap described.

Why D is wrong: Adam can converge faster than plain SGD and may generalise differently, but choosing an optimiser does not intrinsically penalise weight magnitude in the loss function, so it does not directly regularise in the way described.

See more NCA-GENM practice questions, answers explained.

More in this domain

Back to all Core Machine Learning and AI Knowledge objectives, or the NCA-GENM cert hub.

Examworthy is not affiliated with or endorsed by NVIDIA. Original, blueprint-aligned practice material only.