NCA-GENM domain - 20% of the exam

Core Machine Learning and AI Knowledge

Core Machine Learning and AI Knowledge is 20% of the NVIDIA-Certified Associate: Generative AI Multimodal (NCA-GENM) exam. These are the objectives it covers, each with practice questions and worked explanations.

Objectives in this domain

Sample question from this domain

Free sampleCore Machine Learning and AI Knowledgemedium

A pipeline ingests 16-bit greyscale medical images and feeds them into a convolutional network. Before training, each pixel value is divided by 65535 so the resulting tensor contains values in the range 0 to 1. What is the primary reason for this operation?

  • ATo normalise input values so that large pixel magnitudes do not cause unstable gradients during back-propagation, improving training convergence. Correct
  • BTo reduce the spatial resolution of the image so that fewer parameters are needed in the first convolutional layer.
  • CTo convert the integer pixel values into a floating-point representation that is compatible with the network's activation functions and loss computation.
  • DTo apply zero-mean standardisation so that each channel has a mean of zero and a standard deviation of one before entering the network.
Explain why pixel-value normalisation is applied to image tensors before neural-network training. Dividing raw pixel values by the maximum representable value maps all inputs to the closed interval 0 to 1. Neural networks trained on unnormalised inputs with very large magnitudes experience large activation values, which produce large loss gradients and can cause weight updates to overshoot, destabilising training. Bounded inputs keep the gradient signal well-conditioned throughout back-propagation, which is the primary purpose of this pre-processing step.

Why A is correct: Scaling inputs to a small, bounded range keeps weight gradients in a workable magnitude, preventing exploding gradients and accelerating convergence - the canonical motivation for input normalisation.

Why B is wrong: Dividing pixel values by a constant does not alter the spatial dimensions of the tensor; downsampling requires pooling or strided convolutions, not scalar division.

Why C is wrong: Casting to float is a necessary step, but the division itself is not required for the cast; the primary motivation is gradient stability from bounded inputs, not merely the numeric type.

Why D is wrong: Dividing by the maximum value scales to the unit interval but does not produce zero-mean output; standardisation requires subtracting the dataset mean and dividing by its standard deviation.

Other domains in this exam

See also the NCA-GENM cert hub, the study guide, and the cheat sheet.

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