Drag and drop the AI/ML concepts on the left to the correct descriptions on the right.
Drag a concept onto its matching description — or click a concept then click the description.
Identifies deviations from normal network behavior, such as unusual traffic spikes or security threats.
Uses historical data and ML models to forecast future network events, like congestion or device failures.
Translates business intents into network policies and continuously verifies that the network meets those intents.
Trains a model using labeled data to classify or predict outcomes, such as identifying specific types of traffic.
Discovers hidden patterns or clusters in unlabeled data, often used for anomaly detection or traffic profiling.
Optimizes network decisions through trial and error, using rewards to learn optimal actions over time.
Why these pairings
Supervised learning uses labeled data, unsupervised finds hidden patterns, reinforcement learning uses rewards, neural networks mimic brain structure, training data is for learning, and inference is applying the model to new data.
Exam trap
The trap is that candidates may confuse the definitions of supervised learning, unsupervised learning, reinforcement learning, and neural networks. Remember: supervised = labeled data, unsupervised = unlabeled patterns, reinforcement = rewards, neural networks = brain-inspired architecture. Focus on the key differentiator: the presence or absence of labels.