Q80 — AWS AIF-C01 Ch.2

Question 80 of 100 | ← Chapter 2

An ML engineer has trained a deep neural network model on a large dataset. What is the process called when using the trained model to make predictions on new, previously unseen data?

• A. Training
• B. Validation
• C. Inference ✓
• D. Feature Engineering

Correct Answer: C. Inference

Explanation

A. Training refers to the process of optimizing model weights and hyperparameters using labeled training data to minimize a loss function — this step is already complete. B. Validation occurs during training to assess model performance on held-out data and prevent overfitting — it is part of the training phase. C. Inference is the process of using a trained model to generate predictions or decisions on new, unseen data — precisely what the ML engineer performs after training a deep neural network. D. Feature engineering involves extracting and transforming relevant features from raw data prior to model training. Thus, applying a trained deep neural network to new, unseen data is called inference — a critical production step often optimized for low latency and efficiency.