Artificial Intelligence
Course Overview
Teachers
Michael Ashcroft
(lectures, examiner)
Office 1320, tel: 018 - 471 33 90,
Email: 
M. Andreina Francisco (assistant)
Office 1345A, tel: 018 - 471 10 55,
Email:
Language
The course will be given in English. Assignments and presentations must be in English.
Lab assignments
The course includes two lab assignments. Lab assignments should be solved in groups of four students. For each lab assignment, you can select one task from the given options:
- Lab 1:
- Implement a planner capable of solver PDDL problems.
- Generate search or dynamic programming routines that you can use to compete on the 'Delivery Man' game.
The full documentation for Lab 1 is available here
. Files for Lab 1 are available here
or here (debug build). This includes the libraries and header files for the Delivery Man game as well as a exe file that will run my basic Delivery Man implementation in a GUI so you can get a visual feel for the problem. It also contains source code for a JSON parser that you may use (with no guarantees) for the Planner/Scheduler task. An explanation of the entry points in the DeliveryManClient library is given here.
- Lab 2:
- Implement a Hidden Markov Model based routine that you can use to compete on the 'Where's Croc' game.
- Implement a Markov Random Field based handwriting recognition application
- EXTERNAL PROJECTS Produce a graphical model of a system from expert knowledge and perform analysis regarding the causal influences on a chosen variable. (See discussion on external projects below.)
The full documentation for Lab 2 is available here. Files for Lab 2 are available here or here (debug build). This includes the libraries and header files for the Where's Croc game as well as a exe file that will run a basic console version of the game if you wish to get a feel for it in this way.
Assignments handed in before their respective deadlines (see table below) will be marked and returned as soon as possible. All assignments must be passed in order to pass the lab component of the course.
Note that one of the requirements for a top grade in the course as a whole is that the assignments (serious attempts) have been handed in before their respective deadlines. Assignments handed in after the last scheduled day will be accepted, but may not be marked until the end of the semester (Aug 31). Students who fail to hand in the reports before the end of the semester, fail the course.
Groups undertaking external projects may need altered deadlines. See discussion on external projects below.)
| Assignment | Deadline |
|---|---|
| 1 | 23rd September |
| 2 | 12th October |
External Projects (Assignment 2)
The basic task specifications and details of the external partners provide information required for deciding upon an assignment and contacting the external partners. Addition instructions may be provided closer to commencement.
Do not contact the external partners unless your group has been confirmed as undertaking the given project.
NB If you cannot view these pdf documents in your browser, try and download them to your local machine and use Adobe Reader.
| Assignment | Documents |
|---|---|
| Gender and Pay | Task Overview and Contact Details |
| Classroom Teaching | Task Overview and Contact Details |
| Greenhouse Cultivation | Task Overview and Contact Details |
Presentation
In addition to the lab assignments, the course includes a presentation assignment which should also be done in groups of two students. Groups should select one of the topics below:
- Investigate the potential application to the focus area of 'Smart Cities' of a particular technique or class of techniques covered in the course.
- Discuss current state-of-the-art for a particular technique covered in the course. Include a review of current literature.
Examination
Examination in this course is a sum of three parts; The lab assignments, the presentation and a written exam, in total worth 5 credits.
The presentation is graded U (fail) or G (pass). The lab assignments are graded U, 3, 4 or 5. Together (indivisibly) the lab projects and presentation are worth 2 credits.
The written exam is graded U, 3, 4 or 5 and is worth 3 credits.
The total grade on a complete course is the average grade of the lab projects (weighted at 25% each) and the exam (weighted at 50%), rounded to the closest integer. In addition you must pass all the labs and, for grade 5 specifically, all deadlines during the course must have been met:
| Presentation | Lab Projects (Total) | Exam | Additional requirements | Total |
|---|---|---|---|---|
| Pass | 6 | 3 | 3 | |
| Pass | 7 | 3 | 3 | |
| Pass | 6 | 4 | 4 | |
| Pass | 6 | 5 | 4 | |
| Pass | 7 | 4 | 4 | |
| Pass | 7 | 5 | 4 | |
| Pass | 8 | 3 | 4 | |
| Pass | 8 | 4 | 4 | |
| Pass | 9 | 3 | 4 | |
| Pass | 10 | 3 | 4 | |
| Pass | 9 | 4 | 4 | |
| Pass | 8 | 5 | all deadlines met | 5 |
| Pass | 9 | 5 | all deadlines met | 5 |
| Pass | 10 | 4 | all deadlines met | 5 |
| Pass | 10 | 5 | all deadlines met | 5 |
Old Exams
The exam from last year is available here.
Lectures
NB The schedule may change. The below is as of publication only and you should check the official schedule for changes.
It is expected that you research the topics presented in the lectures further. It is not compulsory that you use the readings specified: There are many fine sources available for free on the internet.
The numbers in the reading column refer to chapters and section in either Russell and Norvig, Artificial intelligence : A Modern Approach, 3rd Ed ('AI') or Haste et al, The Elements of Statistical Learning: Data Mining, Inference and Prediction, 2nd Edition ('SL'). Note that SL is available online as a pdf for free.
The lecture slides will be posted as pdf documents. If you cannot view these in your browser, try and download them to your local machine and use Adobe Reader.
| Week | Day | Time | Room | Type | Topic | Readings | Slides |
|---|---|---|---|---|---|---|---|
| w36 | Thu 2013-09-05 | 13:15 - 15:00 | 1311, ITC | Lecture | Introduction. Discussion regarding what Artificial Intelligence is and its relation to other fields and technology. The relation of this course to others at the University. The idea behind the selection of specific areas for inclusion in this course. | - | Lecture 1 |
| w36 | Fri 2013-09-06 | 13:15 - 15:00 | 1311, ITC | Lecture | Search and Dynamic Programming. Brief discussion on basic search algorithms. Longer discussion on local search algorithms and dynamic programming for path finding. | AI: 3.3,3.4,4.1 | Lecture 2 |
| w37 | Tue 2013-09-10 | 10:15 - 12:00 | 1211, ITC | Lecture | Heuristic search and minimax pruning. Guest lecture by Roland Bol. | AI: 3.5,3.6,5.1-5.3 | Lecture 3 (Part 1),Lecture 3 (Part 2) |
| w37 | Thu 2013-09-12 | 10:15 - 12:00 | 1311, ITC | Lecture | Planning and Scheduling. Overview of Planning Domain Definition Language (PDDL). Forward and backward chaining, and partial order and graph planning algorithms. Discussion of the first lab assignment. | AI: 10.1-10.3,10.4.4,11.1 | Lecture 4 - Planning or as pptx, Lecture 4 - Scheduling |
| w37 | Fri 2013-09-13 | 13:15 - 17:00 | 1515, ITC | Lab | First Lab Assignment | - | |
| w38 | Tue 2013-09-17 | 08:15 - 10:00 | 1311, ITC | Lecture | Natural Computation. Guest Lecture by Olle Gällmo | - | Lecture 5 |
| w39 | Wed 2013-09-25 | 10:15 - 12:00 | 1311, ITC | Lecture | Supervised Learning: What are systems? Linear regression and nearest neighbor techniques. Basis changes, splines and wavelets. Kernal methods. Bayesian methods. | SL: 2.3,3.1-3.2,5.1,13.3 | Lecture 6 |
| w39 | Thu 2013-09-26 | 13:15 - 15:00 | 1211, ITC | Lecture | Things Markov: Markov Chains. Hidden Markov Models and the Viterbi algorithm. Monte Carlo methods (Gibbs Sampling, which is a version of Markov Chain Monte Carlo). Markov Random Fields and image recognition. | AI: 15.3, SL: 17.1-17.2 | Lecture 7 |
| w40 | Mon 2013-09-30 | 13:15 - 15:00 | 1311, ITC | Lecture | Graphical models. Functional and stochastic models. Dynamic and state models. Parameter estimation. Pseudo-experimental analysis from causal networks. Discussion of the second lab assignment. | AI: 14.1-14.5,15.4-15.5 | Lecture 8
|
| w40 | Tue 2013-10-01 | 13:15 - 17:00 | 2510, ITC | Lab | Second Lab Assignment | - | |
| w41 | Mon 2013-10-07 | 08:15 - 10:00 | 1311, ITC | Lecture | Unsupervised Learning: Clustering, associative rules and principle component analysis. | SL: 14.2-14.3 | Lecture 9
|
| w41 | Wed 2013-10-09 | 15:15 - 17:00 | 1311, ITC | Lecture | AI in computer games. Guest lecture by Olle Gällmo | - | Lecture 10
|
| w41 | Fri 2013-10-11 | 13:15 - 15:00 | 1311, ITC | Lecture | Natural language analysis. Guest lecture. TBA | - | |
| w42 | Mon 2013-10-14 | 13:15 - 15:00 | 1311, ITC | Lecture | Wrap up. Questions for the exam. Discussion of the presentation. | - | Lecture 12
|
| w42 | Tue 2013-10-15 | 13:15 - 17:00 | 1213, ITC | Presentations | Presentation Assignment (Need to book a time. Information sent by email) | - | |
| w43 | Mon 2013-10-21 | 08:00 - 17:00 | TBA | Exam | - | - |