Artificial Intelligence (AI) - old course, valid until 2020/2021

You can “subscribe” to any page on the wiki and you’ll get an email message whenever the page has changed. To do that, click the Manage Subscriptions link.

• The course presentation form.

• The official course page in OLE.

The official week-by-week Faculty timetable can be found on the RIS of the course. Note that sometimes a LAB may be transformed into a LECTURE and vice-versa.

Office hours: anytime, by previous appointment by email to the lecturer (Enrico Franconi). In any case the lecturer is always available for the period after any lecture.

Lectures are offered online using Microsoft Teams. There is a video on how to start with Teams.

• Exclusively English.

Online lectures are video-recorded for the convenience of students. Students should switch off their video-camera. In any case, participation to a video-recorded online lecture indicates the student's consent to be included in the recording of the online lecture. Students can watch and use the the video-recorded online lectures only for learning purposes and for their exclusive personal use. Any other use, such as reproduction, distribution or sharing of the didactic material, is illegal.

All information concerning the personal data processing can be found under the following link: https://guide.unibz.it/en/covid-19/.

• Main book: David Poole and Alan Mackworth. Artificial Intelligence: Foundations of Computational Agents. Cambridge University Press, 2010, 2nd edition 2017.
• Auxiliary book: Stuart Jonathan Russell and Peter Norvig. Artificial Intelligence: A Modern Approach. Prentice Hall, 4th edition 2020.
• Reading list from the UniBZ Library: 76212_19-20-2_CS Artificial Intelligence

Below you can download the relevant chapters of the book used in the various parts of the course.

The following is the standard material, it may be adjusted during the course.

• Slides: Welcome Aboard
• Slides: Artificial Intelligence and Agents part 1, part 2, part 3, part 4
  • Material: chapter 1 of Poole and Mackworth
• Slides: Agent Architectures and Hierarchical Control part 1, part 2
  • Material: chapter 2 of Poole and Mackworth
• Slides: States and Searching part 1, part 2, part 3, part 4
  • Material chapter 3 of Poole and Mackworth
• Slides: Features and Constraints part 1, part 2,
  • Material chapter 4 of Poole and Mackworth
• Slides: Propositions and Inference part 1, part 2, part 3, part 4
  Slides: Diagnosis part 1, part 2, part 3, part 4, part 5
  • Material chapter 5 of Poole and Mackworth
• Slides: Supervised learning and Decision Trees: part 1, part 2, part 3.
  • Material chapter 7 of Poole and Mackworth.
• Slides: Individuals and Relations part 1, part 2, part 3, part 4
  • Material chapter 12 of Poole and Mackworth
• Slides: Natural Language Understanding
  • Material chapter 12 of Poole and Mackworth
• Slides: Introduction to Machine Learning

Start date: XXX March 2021

• LAB 1: Robot Control (applet).
  • Quick start
  • General help
  • Tutorial 1 with video
  • Tutorial 2 with video
  • Tutorial 3 with video
• LAB 2: Graph Searching with uninformed techniques (applet)
  • Manuals:
    • Quick start
    • General Help
    • Tutorial 1: Creating a New Graph with video
    • Tutorial 2: Loading a Preexisting Graph
    • Tutorial 3: Solving a Graph with video
    • Tutorial 4: Search Options
  • Explore with the search applet the Delivery Robot (Acyclic), the Delivery Robot (cyclic), the Vancouver Neighbourood, the Module 4 sample problem graphs: with Depth First, Breadth First, Lowest Cost First search strategies using different Neighbour Ordering Strategies; practice also with the quiz facility.
  • Do the Practice Exercise 3.B.
  • Exercise: practicing different search strategies with the graph in this slides (solution as a XML file for the applet)
  • Create your own problem graph for a delivery robot starting from a map with edge costs.
  • Create a problem graph for a simple problem chosen by you.
• LAB 3: Graph Searching with Heuristics
  • Explore the Delivery Robot (Acyclic), the Delivery Robot (cyclic), the Vancouver Neighbourood, the Module 4 sample problem graphs: with Best First, Heuristic Depth First, A*, Branch and Bound search strategies, with or without Multiple-Path Pruning or Loop Detection, using different Neighbour Ordering Strategies. Explore also the behaviour with the abovementioned search graphs with potentially non terminating depth first strategies (e.g., Depth First or Heuristic Depth First) without cycle checking, and with loop detection or multiple path pruning.
  • Do the Practice Exercises 3.C, 3.D, 3.E.
  • Exercise: 3.4 (solution).
• LAB 4: Constraints - Consistency
  • Consistency Based CSP Solver (applet):
  • Do the Practice Exercises 4.A, 4.B.
  • Exercises: 4.2, 4.3 (only a,b), 4.5 - CSP and arc consistency (solutions).
• LAB 5: Propositions and Inference
  • Getting started with AILog2, a representation and reasoning system for definite clauses, with declarative debugging tools.
    • Download the file ailog2.pl, install and launch SWI Prolog, and load (consult) AILog2 in Prolog:
      • Windows: ?- consult("C:\\path-to-file\\ailog2.pl").
      • Mac: ?- consult('/path-to-file/ailog2.pl').
    • Go through the AiLog2 manual, from Section 1 to Section 6.
      • To load a knowledge base file from AILog2:
        • Windows: ailog: load 'C:\\path-to-kbfile\\kbfile.ail'.
        • Mac: ailog: load '/path-to-kbfile/kbfile.ail'.
  • Play with the elect_prop.ail electrical wiring example 5.7 from Section 5.3 of the book and the slides.
  • Do Exercises 5.1, 5.2 (not question b), 5.3, 5.4 (solutions)
  • Find other AILog knowledge base examples here

* LAB 6: Debug, Diagnosis, Abduction * Keep using AILog2 * Go through Sections 6, 7, 9 of the manual * Play with the following knowledge bases: * elect_ask.ail electrical wiring example with askables; Example 5.10 from Section 5.3.2 * elect_bug.ail the buggy electrical wiring knowledge base from Example 5.14 in Section 5.3.4.1 * elect_bug2.ail the buggy electrical wiring example from Exercise 5.6 * elect_bug3.ail a buggy electrical wiring example, which fails to prove lit_l2, but should succeed * elect_cbd.ail electrical wiring example for consistency-based diagnosis; Example 5.20 in Section 5.4.3 * elect_abd.ail electrical wiring example with abduction; Example 5.31 in Section 5.6 * Exercises: 5.5, 5.6, 5.7, 5.9, 5.13 /*(solutions)*/

• LAB 7: Individuals and Relations
  • Keep using AILog2
  • Go through Section 8 of the manual
  • Play with the relational representation of the electrical environment and with the AM and PM time representations
  • Do the following exercises (you may use the AILog2 KBs listed in the resources for Chapter 12):
    • 12.3 (bottom-up derivation)
    • 12.4,12.5,12.6 (top-down derivation) with the knowledge base about rooms
    • 12.8,12.9,12.14 (SLD-resolution with functions)
    • (solutions)

• LAB 8: Natural Language Processing
  • Understand and play with the knowledge bases discussed in the lecture:
    • cfg_simple.ail a simple context free grammar and associated dictionary from Figures 12.6 and 12.7 in Section 12.5 (also in Prolog)
    • nl_numbera.ail a simple grammar that enforces number agreement and builds a parse tree; from Figure 12.9 in Section 12.6.5 (also in Prolog)
    • nl_interface.ail a simple natural language interface to a database; from Figures 12.10 and 12.11 in Section 12.6.6 (also in Prolog)
  • Exercise: 12.17 (solution) - Following the code of the simple natural language interface to a database, construct a knowledge base with a grammar (where a relative clause could be treated similarly to a prepositional phrase) and a dictionary to find all the answers to the following geographical questions:
    • Which is a country?
    • Which is a Spanish speaking country?
    • Which is the capital of Chile?
    • Which is the capital of a large country?
    • Which is a country that borders Chile?
    • Which is a country that borders a country that borders Chile?
    • Which is the capital of a country that borders Chile?
    • Which country borders Chile?
    • Which country that borders Chile borders Paraguay?
  • using the following facts:

% country(C) is true if C is a country
country(argentina).
country(brazil).
country(chile).
country(paraguay).
country(peru).

% large(C) is true if the area of C is greater than 2m km^2
large(brazil).
large(argentina).

% language(L) is true if L is a language
language(spanish).
langauge(portugese).

% speaks(Country,Lang) is true of Lang is an official language of Country
speaks(argentina,spanish).
speaks(brazil,portugese).
speaks(chile,spanish).
speaks(paraguay,spanish).
speaks(peru,spanish).

capital(argentina,'Buenos Aires').
capital(chile,'Santiago').
capital(peru,'Lima').
capital(brazil,'Brasilia').
capital(paraguay,'Asunción').

% borders(C1,C2) is true if country C1 borders country C2
borders(peru,chile).
borders(chile,peru).
borders(argentina,chile).
borders(chile,argentina).
borders(brazil,peru).
borders(peru,brazil).
borders(argentina,brazil).
borders(brazil,argentina).
borders(brazil,paraguay).
borders(paraguay,brazil).
borders(argentina,paraguay).
borders(paraguay,argentina).

• LAB: Decision Trees
  • Answer the following questions:
    • What does an arc represent in a decision tree?
    • What does a non-leaf node represent in a decision tree?
    • What does a leaf node represent in a decision tree?
  • Do exercises from (slides)
  • If-time: do Exercise 7.3
• More KBs:
  • AILog exercises (use extensively How?, Why not?, Why? questions and Depth Bound):
  • elect_relational.ail electrical wiring example from Section 12.3.2
  • west.ail a knowledge base about rooms from Figure 12.2 in Section 12.3.3
  • before.ail the before relation of Example 12.26 in Section 12.5
  • trees.ail the tree code of Example 12.27 in Section 12.5
  • elect_cbd.ail electrical wiring example for consistency-based diagnosis; Example 5.20 in Section 5.4.3
  • elect_naf.ail electrical wiring example with negation as failure; Example 5.26 in Section 5.5
  • beach.ail default reasoning about swimming at beaches; Example 5.27 in Section 5.5.1
  • bronchitis.ail diagnosis Example 5.30 in Section 5.6
  • elect_abd.ail electrical wiring example with abduction; Example 5.31 in Section 5.6
• LAB: Recap of Prolog. Material here.
• LAB: More on Graph Searching (III), and Programming Search in Prolog
  • Exercise: do 3.3(d) using A* with multiple-path pruning search strategies.
    • Test your solution: External Link
  • Do Practice Exercise 3.C.
  • Do Practice Exercise 3.D.
  • Practise with iterative deepening and branch-and-bound
    • Check Example 3.19 on branch-and-bound search
  • Search in Prolog with SWI Prolog or XSB with the InterProlog Studio IDE.
  • Another (simpler?) version in prolog of Depth-first and Breadth-first
  • A prolog implementation of the generic search algorithm (by Davide Lanti) External Link

The exam paper will contain problem solving questions, where you have to apply the techniques and theory studied during the course to concrete examples. There will be no pure theory question. The questions will be mirroring the exercises done in the lab and exercises from the main course book. Answers to questions should be formulated precisely and completely, of course. Regarding the topic that have been done during the lab using simulation software, the written answer in the exam paper should contain the information relevant to the question as it would be provided by the simulation, and as it is discussed in the book and during the lectures. No “mock” exam paper is provided, given the quite simplifying specifications given above.

• Requirements: (1) a computer with a webcam and a microphone, (2) a smartphone with camera and internet connection, (3) sheets of paper to write the answers to the questions.

• Platform to communicate with the professor during the written exam (login with your UniBZ account): MS Teams at this MS Teams link (the link will be updated on the day of the exam). Keep your video camera off in MS Teams (otherwise it wouldn't work in OWLTS). Connect to MS Teams five minutes before the start of the exam, and keep the connection open during the whole duration of the exam.
  In case of technical problems, you can ask for help via chat using MS Teams with the faculty secretariat (MS Teams link for Christine Kirchler or MS Teams link for Federica Cumer) or with the ICT technicians (MS Teams link for Denise Delai, MS Teams link for Cristiano Cumer).

• Platform for the written exam (login with your UniBZ account): OWLTS at https://unibz.owlts.com .

• A smartphone to take photos of the written sheets used to answer the questions. Make sure that the photos are contained in the right order within a single PDF document. The PDF document will be uploaded to OWLTS from the computer. IMPORTANT: Practice in advance on taking photos with the phone, on generating a single PDF file from the photos in the right order, on moving files from the phone to the computer.

This document explains the technical details of working with OWLTS.

Please check that the abovementioned platforms do work before the exam; in case of any issue, please contact ICT in advance.

• For the duration of the exam, the video camera integrated in the PC or notebook must be kept switched on.
• Microphones can be muted but have to be open at any time upon the commission request.
• The entire exam is recorded.
• For the duration of the exam, you are not allowed to access to the internet.
• The exam is weakly “open book”, in the sense that you can consult your own notes or books, but you cannot consult computer- or internet-based material.
• Only blank sheets of paper, a pen, and the abovementioned notes and books may be used as aids.
• You should position the video camera in a way the desk, the hands, the sheets of paper, and the notes and books are continuously visible.
• During the exam, you may not use a mobile phone, or any other internet or technical device, except to request technical assistance.
• Mobile phones have to be out of reach till the end of the exam. Only at the end, the phone will be used to take photos of the written sheets used to answer the questions.
• All the windows on your computer - except Chrome - must be closed, including all the Chrome Tabs. Only the exam page can remain open.
• You are not allowed to stand up or interact with third parties during the entire examination.
• If you do not follow the instructions, do not follow the rules listed here or if you try to use external help, you will be excluded from the exam.
• At the beginning of the exam, connect to MS Teams. If you need to ask questions to your professor, please use MS Teams, as instructed in the web page of the course.

I acknowledge that University examinations require all students to respect the highest standards of academic integrity. For the examination I am about to take, I make the following pledge:

• all the work will be my own, and I will not plagiarize from any source;
• I will not obtain or seek to obtain an unfair advantage by communicating or attempting to communicate with any other person during the examination; neither will I give or attempt to give assistance to another student taking the examination;
• I will stop writing immediately at the designated end time of the examination and will make no modification to my script thereafter.

I understand that, if suspected of violating this pledge, I will be liable to be referred to the Organ competent for disciplinary issues and may be subject to disciplinary action. I am also aware that voice- or videorecording of the online exams as well as the dissemination of their contents is neither provided for nor allowed.