Prepare and test your skills
Prepare and test your skills
Free sample question, answered and explained below. The Generative AI Leader practice bank has 2,191 more like it. Try 10 free: instant score, no account.
A retail company has gathered a massive collection of customer feedback, purchase histories, and support inquiries that lack explicit categories or labels. The data science team wants to explore natural groupings within the data to segment customers and discover underlying themes before using the raw text to pre-train a foundation model.
Which machine learning approach is designed to discover inherent patterns and structures in unlabeled data?
Reinforcement Learning from Human Feedback (RLHF) is an advanced alignment technique used to fine-tune pre-trained large language models. It uses human preference ratings to train a reward model, which then optimizes the generative model's policy to produce helpful, honest, and harmless responses.
RLHF represents the final alignment stage in the modern generative AI pipeline, whereas unsupervised learning underpins the initial exploration and pre-training phases.
Unsupervised learning is a foundational branch of machine learning where algorithms analyze and discover inherent structures, patterns, and groupings directly from unlabeled data without human guidance or predefined target outcomes.
Unsupervised learning is the exact approach designed for pattern discovery across unstructured, unlabeled data. It enables both exploratory segmentation and the initial self-supervised pre-training needed for large foundation models.
Reinforcement learning (RL) is a machine learning paradigm where an autonomous agent learns to make sequential decisions by interacting with an environment. The agent learns through trial and error, receiving numerical rewards for desirable actions and penalties for undesirable actions to maximize cumulative reward over time.
Reinforcement learning principles are applied to generative AI models via post-training alignment, but not for discovering latent clusters in raw data repositories.
Supervised learning is a machine learning methodology where algorithms are trained on labeled datasets. In this paradigm, every training example consists of an input paired with a corresponding target output or ground-truth label (such as spam/not spam or categorized support tickets).
While supervised fine-tuning is vital in later stages of foundation model customization, it relies on structured prompt-response pairs rather than raw, unlabeled corpora.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated