CS 480 Final Exam Terms and Concepts
Chapter 1
- What is AI?
- thinking vs acting, humanly vs rationally
- a little bit of history
Chapter 2 Intelligent Agents
- Terms: agent, environment, sensors, actuators, percept, agent
function, agent program, rational agent, performance measure,
information gathering, learning
- Describing task environments (PEAS)
- Properties of task environments:
- fully observable vs partially observable
- single agent vs multiagent
- deterministic vs stochastic
- episodic vs sequential
- static vs dynamic
- discrete vs continuous
- kinds of agent programs:
- table-driven agents
- simple reflex agents
- model-based reflex agents
- goal-based agents
- utility-based agents
- learning agents
- representational complexity: atomic, factored, structured
Chapter 3 Search
- problem-solving agents
- formulating problems, consisting of:
- initial state
- actions
- transition model
- goal test
- path cost function
- basic tree search and graph search algorithms
- measuring performance:
- completeness
- optimality
- time complexity
- space complexity
- Uninformed search strategies:
- breadth-first search
- uniform-cost search
- depth-first search
- depth-limited search
- iterative deepening depth-first search
- Informed search strategies:
- greedy best-first search
- A* search
- what heuristics are, where they come from, what makes a
heuristic admissible
Chapter 4 Local Search and Optimization
- hill climbing search
- simulated annealing
- genetic algorithms
Chapter 5 Adversarial Search
- basic organization of adversarial search: initial state,
transition model, terminal test, utility function
- minimax algorithm
- minimax with alphabeta pruning
- effect of adding chance to adversarial search
Chapter 6 Constraint Satisfaction Problems
- Basic idea that a CSP consists of a set of variables, domains
for each variable and a set of constraints on the values of the
variables.
- Basic idea of the backtracking search algorithm.
- Some notion of the heuristics for doing effective search.
- Idea of using local search (on complete problem instances) to
solve CSP problems
Chapter 7 Propositional Logic
- terms: knowledge base, sentence, inference, entailment,
syntax, semantics
- knowledge level, logical level, implementation level
- issues that arise in the Wumpus World
- understand and be able to explain Figure 7.6
- soundness and completeness
- converting to CNF and doing resolution in propositional logic
Chapter 8 and a bit from 9 First Order Logic
- terms: objects, relations, functions, quantifiers
- expressing simple English sentences in FOL
Chapter 10 Classic Planning
- states, actions and goals
- preconditions and effects (add list and delete list)
- partial-order planning, search through plan space
Chapter 13 Uncertainty
- reasons why uncertainty exists
- Boolean random variables, discrete random variables
- prior probability, probability distribution, full joint probability distribution
- product rule P(a and b) = P(a|b)P(b) = P(b|a)P(a)
- independence (absolute and conditional)
- Bayes' Rule
Chapter 18 Inductive Learning
- kinds of learning: supervised, unsupervised, reinforcement
- terms: induction, Occam's razor
- learning decision trees
- basic idea of how it works
- training set vs. test set
- how attributes are chosen
Chapter 19 Neural Networks and Deep Learning
- basic model of a neuron, what the parts are
- terms: activation, units, weight, bias weight, threshold function, feed-forward vs. recurrent networks, hidden units, linearly separable, back-propagation algorithm, cluster analysis
- Build simple networks "by hand"
- What tensors are and why they are relevant
- Overfitting vs. underfitting in a network. A little bit about
what can be done about each of them.
- Convolutional Networks. What they are. What they are good for.
Section 23.5 Speech Recognition
- terms: speech recognition, speech understanding, phones, coarticulation, bigram model, hidden Markov models
Chapter 22 and a bit of 23 Communication (Language)
- terms: natural language vs. programming language, grammar, formal language, parse trees, syntax, semantics, speech acts
- stages in communication and ways they can go wrong
- issues in "real" language
- ambiguity (be able to give different kinds of examples)
- indexicality, metonymy, metaphor, noncompositionality
Chapter 25 Philosophical Foundations
- weak vs. strong AI
- thought experiments: Searle's Chinese room, Turing test