AI Advanced 1

Which of the following definitions best captures the multifaceted nature of AI as discussed in the lectures?

• AI as purely rational decision-making, ignoring human cognition

• AI as systems that can think like humans, act like humans, think rationally, and act rationally

• AI as heuristic-based problem solvers without regard to intelligence

• AI as pattern-matching systems confined to rule-based logic

Consider the humorous definition of AI: “things in Computer Science we don’t yet know how to do.” How does this perspective influence research directions in AI?

• It discourages exploring new problems since they are always unsolved

• It emphasizes that once a problem is solved, it’s no longer considered AI, driving researchers to explore frontier challenges

• It encourages focusing only on well-known solutions rather than new problems

• It implies that AI will remain static once certain tasks are achieved

From the foundational perspectives mentioned, if an AI system’s approach is to infer and predict human thought processes through psychological experiments, which approach is it aligned with?

• Acting rationally

• Thinking like humans

• Acting like humans

• Thinking rationally

Some early philosophical inquiries into intelligence treated human cognition as mechanistic. What key impact did these early philosophical views have on AI’s development?

• They dismissed formal logic as a tool for modeling thought

• They laid conceptual groundwork that human thought could be simulated or emulated by machines

• They concluded machine intelligence was impossible

• They required abandoning quantitative methods in AI

How did mathematics contribute to the development of AI, according to the lecture material?

• By providing only data structures for memory management

• Through formalizing computation, enabling complexity analysis, and offering tools like logic and probability

• By solely proving that all problems are tractable with enough time

• By avoiding probabilistic reasoning in favor of deterministic models

AI’s role in computer engineering has been historically significant. Which of the following best demonstrates this relationship?

• AI borrowed all concepts from computer engineering but contributed nothing in return

• AI ideas influenced developments like time-sharing and object-oriented programming, shaping computer science’s evolution

• Computer engineering remained unaffected by AI advances in algorithms and architectures

• AI and computer engineering developed entirely in isolation

In major AI areas, search techniques are considered fundamental because:

• They solve only small puzzles but not complex, real-world problems

• Most AI problems can be reformulated as finding a path through a space of states or hypotheses

• They replace the need for machine learning and logic entirely

• They ensure solutions without requiring heuristics or optimization

The concept of a rational agent is central in AI. What key characteristic defines a rational agent in this context?

• An agent that always follows human intuition rather than evidence-based reasoning

• An agent that selects actions to maximize expected utility given its knowledge and environment

• An agent that never uses probabilistic methods

• An agent that ignores percepts and acts randomly

Which scenario best illustrates the “acting like humans” facet of AI research?

• A chess engine using pure logic without mimicking any human strategies

• A voice assistant that understands natural language and responds as a human would in conversation

• A pathfinding algorithm that seeks the optimal route but doesn’t communicate

• A theorem prover that uses formal logic but no imitation of human cognitive steps

Natural Language Processing (NLP) in AI focuses on enabling machines to:

• Interpret, understand, and generate human language in a meaningful, context-aware manner

• Translate only formal languages like programming code

• Ignore semantics and focus solely on syntax patterns

• Replace all forms of knowledge representation with raw text manipulation