Fundamentals of AI and ML cheat sheet
All 8 AIF-C01 flashcards for Fundamentals of AI and ML on one page, grouped by exam objective. Print it or save it as a PDF for a last-minute read.
Explain basic AI concepts and terminologies
- Supervised vs unsupervised vs reinforcement learning?
- Supervised learns from labelled examples. Unsupervised finds patterns in unlabelled data. Reinforcement learns by trial and reward towards a goal.
- Labelled vs unlabelled data?
- Labelled data has the right answer attached, such as "fraud" or "not fraud". Unlabelled data is raw, with no answers.
- Structured vs unstructured data?
- Structured data fits rows and columns, like a sales table. Unstructured data doesn't: documents, audio and video.
- Batch vs asynchronous inference?
- Batch scores a whole dataset offline, with no endpoint left running. Asynchronous queues single large or slow requests.
Identify practical use cases for AI
- When is AI the wrong choice?
- When the rules are known and the answer must be exact, or when the cost outweighs the benefit. Plain logic is cheaper and easier to audit.
- Traditional ML or a foundation model: when does traditional ML fit better?
- A narrow prediction on structured, labelled data, such as a table of sales. Foundation models suit open-ended content and often misread tables.
Describe the AI/ML development lifecycle
- Precision vs recall?
- Precision: of everything flagged, how many were right. Recall: of all real cases, how many were caught.
- Why monitor a model after deployment?
- Real-world data changes over time (drift), so performance slips. Monitoring catches it early and signals when to retrain.