Artificial Intelligence Mastery Quiz

10:00

Search, knowledge representation, generative AI, vision, language and agents. · 10 questions

Question 1 of 10

On an unweighted graph, which search always returns a path with the fewest steps?

aDepth-first searchbGreedy best-first searchcBreadth-first searchdRandom restart

Question 2 of 10

In an expert system, which component decides which rules fire and in what order?

aWorking memorybInference enginecKnowledge basedUser interface

Question 3 of 10

Which task suits a generative model rather than a traditional classifier?

aDrafting a personalised apology reply to a complaintbPredicting tomorrow's units soldcFlagging a transaction as fraud or notdGrouping customers into five segments

Question 4 of 10

Which of these is a computer vision task?

aSentiment analysis of reviewsbMachine translationcSpeech-to-text transcriptiondDetecting helmets in CCTV footage

Question 5 of 10

With this heuristic, A* on a weighted graph behaves exactly like which algorithm?

h = lambda node: 0
path, cost = a_star(graph, start, goal, h)

aUniform-cost search (Dijkstra)bDepth-first searchcGreedy best-first searchdBidirectional search

Question 6 of 10

How many facts does `forward_chain` end with here?

RULES = [
    ({"supplies_goods", "turnover_above_40l"}, "must_register_for_gst"),
    ({"must_register_for_gst"}, "files_monthly_returns"),
]
case = {"supplies_goods", "turnover_above_40l"}
print(len(forward_chain(case, RULES)))

a2b3c4dIt loops forever

Question 7 of 10

A bank must explain every rejected loan application to the customer and the regulator. Which approach fits best?

aAn LLM that writes a free-text decision for each applicantbA documented rule set or interpretable model, with the deciding factors loggedcA deep network chosen purely for highest accuracydA clustering model that groups similar applicants

Question 8 of 10

In a tool-calling agent, what happens immediately after the model returns a tool call?

if response.stop_reason == "tool_use":
    result = run_tool(call.name, call.input)
    messages.append({"role": "user", "content": tool_result(call.id, result)})

aThe loop ends and the tool result is shown to the userbThe model executes the function inside itselfcThe tool result replaces the system promptdYour code runs the function and appends its result, then calls the model again

Question 9 of 10

A routing heuristic uses straight-line distance multiplied by 1.5 to allow for detours. Why can A* now return a sub-optimal route?

aMultiplication makes the priority queue slowerbThe heuristic becomes negative for nearby nodescThe heuristic may overestimate the remaining cost, so it is no longer admissibledA* only accepts integer heuristics

Question 10 of 10

Which change most reduces the risk of an agent taking a harmful, irreversible action?

aIncreasing the temperature so it explores more optionsbRequiring human approval for write tools while read-only tools run automaticallycGiving it more tools so it always has an alternativedEnlarging the context window