JCU Logo
JOHN CABOT UNIVERSITY

COURSE CODE: "CS 202"
COURSE NAME: "Artificial Intelligence Concepts"
SEMESTER & YEAR: Fall 2016
SYLLABUS

INSTRUCTOR: John Ewing
EMAIL: [email protected]
 HOURS: TTH 1:30PM 2:45PM
TOTAL NO. OF CONTACT HOURS: 45
CREDITS: 3
PREREQUISITES: Recommended: CS 101
OFFICE HOURS:
COURSE DESCRIPTION:
This course is designed for the general student to provide an INTRODUCTORY overview of artificial intelligence (no computer programming skills are necessary). This course will discuss intelligent agents and the building blocks of artificial intelligence: knowledge bases, reasoning systems, problem solving, heuristic search, machine learning, and planning.
SUMMARY OF COURSE CONTENT:
  • brief history of artificial intelligence
  • intelligent agent models
  • utility functions
  • preference logic
  • categorization of problems (e.g. constraint-satisfaction and optimization)
  • search techniques, including heuristic search
  • knowledge bases and manipulation
  • supervised and unsupervised machine learning
  • reinforcement learning
  • task planning
  • engineering intelligent agents
LEARNING OUTCOMES:
  1. Understanding of how intelligent agents operate and interact with their environment.
  2. Familiarity with state-of-the-art techniques for solving difficult problems.
  3. Knowledge of how machines learn and understanding of critical problems like overfitting.
  4. Preparation for a world where the presence of artificial intelligence is becoming ubiquitous.
  5. Improved ability to leverage artificial intelligence in their lives (both at work and at home).
  6. Greater awareness of which career areas are likely to suffer labor dislocation.
TEXTBOOK:
Book TitleAuthorPublisherISBN numberLibrary Call NumberCommentsFormatLocal BookstoreOnline Purchase
Artificial Intelligence: A Modern Approach (Third Edition)Stuart J. Russell and Peter NorvigPrentice Hall860-1419506989     
REQUIRED RESERVED READING:
NONE

RECOMMENDED RESERVED READING:
Book TitleAuthorPublisherISBN numberLibrary Call NumberComments
Continuous Preference Logic for System EvaluationJ. J. DujmovicIEEE Transactions on Fuzzy Systems vol. 15, no. 6, pp. 1082-1099, Dec. 2007   
GRADING POLICY
-ASSESSMENT METHODS:
AssignmentGuidelinesWeight
Homework (8 assignments)Very brief written assignments (1/2 page to 1 page). 20%
Midterm ExamThis exam will cover concepts from 1st half of semester. A midterm review during previous lecture will give scope of possible questions.25%
Final ExamThis exam will cover concepts from 2nd half of semester. A final exam review during previous lecture will give scope of possible questions.25%
Final ProjectStudents will test an artificial intelligence technology and present their findings to the class. Students will also submit a short write-up of the project (4-5 pages).30%

-ASSESSMENT CRITERIA:
AWork of this quality directly addresses the question or problem raised and provides a coherent argument displaying an extensive knowledge of relevant information or content. This type of work demonstrates the ability to critically evaluate concepts and theory and has an element of novelty and originality. There is clear evidence of a significant amount of reading beyond that required for the course
BThis is highly competent level of performance and directly addresses the question or problem raised.There is a demonstration of some ability to critically evaluatetheory and concepts and relate them to practice. Discussions reflect the student’s own arguments and are not simply a repetition of standard lecture andreference material. The work does not suffer from any major errors or omissions and provides evidence of reading beyond the required assignments.
CThis is an acceptable level of performance and provides answers that are clear but limited, reflecting the information offered in the lectures and reference readings.
DThis level of performances demonstrates that the student lacks a coherent grasp of the material.Important information is omitted and irrelevant points included.In effect, the student has barely done enough to persuade the instructor that s/he should not fail.
FThis work fails to show any knowledge or understanding of the issues raised in the question. Most of the material in the answer is irrelevant.
-ATTENDANCE REQUIREMENTS:
ATTENDANCE REQUIREMENTS AND EXAMINATION POLICY
You cannot make-up a major exam (midterm or final) without the permission of the Dean’s Office. The Dean’s Office will grant such permission only when the absence was caused by a serious impediment, such as a documented illness, hospitalization or death in the immediate family (in which you must attend the funeral) or other situations of similar gravity. Absences due to other meaningful conflicts, such as job interviews, family celebrations, travel difficulties, student misunderstandings or personal convenience, will not be excused. Students who will be absent from a major exam must notify the Dean’s Office prior to that exam. Absences from class due to the observance of a religious holiday will normally be excused. Individual students who will have to miss class to observe a religious holiday should notify the instructor by the end of the Add/Drop period to make prior arrangements for making up any work that will be missed. The final exam period runs until ____________
ACADEMIC HONESTY
As stated in the university catalog, any student who commits an act of academic dishonesty will receive a failing grade on the work in which the dishonesty occurred. In addition, acts of academic dishonesty, irrespective of the weight of the assignment, may result in the student receiving a failing grade in the course. Instances of academic dishonesty will be reported to the Dean of Academic Affairs. A student who is reported twice for academic dishonesty is subject to summary dismissal from the University. In such a case, the Academic Council will then make a recommendation to the President, who will make the final decision.
STUDENTS WITH LEARNING OR OTHER DISABILITIES
John Cabot University does not discriminate on the basis of disability or handicap. Students with approved accommodations must inform their professors at the beginning of the term. Please see the website for the complete policy.
SCHEDULE

Week 1

-Lecture 1:  introduction, course structure, “What is artificial intelligence?” and Turing Test

-Lecture 2:  human brain architecture vs. Von Neumann computer architecture, brief history of AI, recent developments in AI, factors in AI progress; homework 1 assigned

Week 2

-Lecture 1:  intelligent agents, environmental representation, percepts, intelligent agent models

-Lecture 2: setting goals in intelligent agents, measuring performance, predicting performance, utility functions, preference logic; homework 1 due, homework 2 assigned

Week 3

-Lecture 1: types of problems intelligent agents need to solve, constraint satisfaction problems, optimization problems, convexity vs. non-convexity, NP-hard problems

-Lecture 2: intro to problem solving techniques, simplex algorithm, classical search (breadth-first, depth-first, greedy); homework 2 due, homework 3 assigned

Week 4

-Lecture 1:  heuristic search techniques, A* search, local search (hill-climbing, beam search), simulated annealing, exploration versus exploitation

-Lecture 2:  evolutionary algorithms (evolutionary programming, evolution strategies, genetic algorithms); homework 3 due, homework 4 assigned

Week 5

-Lecture 1:  particle swarm, ant-colony optimization, tabu search, no free lunch theorem

-Lecture 2:  lab day #1—solving traveling salesman problem (TSP) with hill-climbing, tabu search, ant-colony optimization, genetic algorithms, and evolutionary programming; homework 4 due, homework 5 assigned

Week 6

-Lecture 1: knowledge bases, logical reasoning, and knowledge manipulation; select final project

-Lecture 2: Bayesian networks, fuzzy logic & hidden Markov models; homework 5 due

Week 7

-Lecture 1:  midterm review

-Lecture 2: midterm exam

Week 8

-Lecture 1: introduction to machine learning, supervised learning, decision trees, k-Nearest Neighbors; homework 6 assigned

-Lecture 2: neural networks, linear discriminants, support-vector machines, kernel methods

Week 9

-Lecture 1: challenges in machine learning, noisy data, overfitting, model regularization, testing, validation, cross-validation; homework 6 due, homework 7 assigned

-Lecture 2: lab day #2—training support-vector machine on sample data-set using grid search and cross-validation to select level of model complexity and regularization coefficient (then replace grid search with genetic algorithm)

Week 10

-Lecture 1: computer vision, deep learning; homework 7 due, homework 8 assigned

-Lecture 2: decision theory, unsupervised learning, reinforcement learning

Week 11

-Lecture 1: overview of planning, planning-graphs, hierarchical task network planning, multi-agent planning; homework 8 due

-Lecture 2: overview of communication and coordination in intelligent agents, human-AI communication, AI-AI communication

Week 12

-Lecture 1: overview of AI engineering, building intelligent agents from AI building blocks, troubleshooting AI

-Lecture 2: areas of active research and development in AI 

Week 13

-Lecture 1: future of AI, philosophical questions, AI’s challenges to humanity (e.g. labor dislocation, AI singularlity)

-Lecture 2: Final Project Presentations

Week 14

-Lecture 1:  Final Project Presentations (final project paper due)

-Lecture 2:  Final Exam review

Week 15

-Final Exam