Intro to AI Concepts Summary & Study Notes

🧭 AI Foundations – AI is about making computers more useful by simulating aspects of intelligence. Different definitions emphasize thinking vs. acting, and humans vs. rationality. The field bridges theory and working programs to yield practical results.

🔎 What is AI? Four common definitions – AI can be seen as: (1) systems that think humans, (2) systems that think rationally, (3) systems that act humans, or (4) systems that act rationally. These lenses guide what we build and how we evaluate success.

💡 Approaches to AI – The four classical approaches are: Acting Humanly, Thinking Humanly, Thinking Rationally, and Acting Rationally. These paths shape research goals and evaluation methods. A key practical idea is that the system should behave intelligently, whether by imitating humans or by achieving goals.

🧠 Rationality & Agents – A rational agent does the right thing to maximize its expected utility. We often express this with a utility function $U(s)$ over states to guide decision making. The basic idea is to select actions that maximize future outcomes given beliefs and observations.

🧰 PEAS Framework – A task description uses PEAS: Performance measure $P$, Environment $E$, Actuators $A$, Sensors $S$. This helps specify what an AI system should optimize, where it acts, and how it perceives the world.

🌍 Environment & Observability – Environments can be fully observable or partially observable, deterministic or stochastic, episodic or sequential, static or dynamic, discrete or continuous, and single-agent or multi-agent. These properties determine what kind of reasoning and algorithms are needed.

🧬 Agent Types & Learning – Agents range from Simple reflex to Reflex with state, Goal-based, Utility-based, and Learning agents. Each type adds more sophistication and memory to cope with complexity and uncertainty.

🕰️ Foundations & History – AI has roots in philosophy, mathematics, economics, neuroscience, psychology, and linguistics. The Dartmouth Conference of 1956 is often cited as the birth moment of CS-based AI research.

💬 Summary – AI is a multidisciplinary effort to create systems that perceive, reason, and act with intelligence. By choosing a lens (think vs. act, human vs. rational), and by modeling the task with PEAS, we can design and evaluate intelligent agents.