Inteligência Artificial

Knowledge of procedural programming.

The course unit is taught through theoretical-practical classes.

The theoretical contents are addressed through the presentation and discussion of fundamental concepts, complemented by case studies, practical examples, and the analysis of Artificial Intelligence applications in different contexts, with particular emphasis on industrial applications.

The practical component focuses on the implementation and experimentation of algorithms, problem solving, and the supervision of practical assignments, promoting the application of the knowledge acquired throughout the course unit.

Upon completion of this course unit, students should be able to:

• Identify and characterize the main paradigms and approaches of Artificial Intelligence.
• Recognize and describe different types and architectures of intelligent agents, understanding their behavior and suitability for different environments and problems.
• Identify and analyze Artificial Intelligence applications in industrial contexts, particularly in production planning and scheduling, industrial process optimization, predictive maintenance, quality control and inspection, logistics, transportation, resource management, and decision support.
• Analyze optimization problems by identifying their main characteristics, variables, constraints, and objectives.
• Formulate Artificial Intelligence problems by identifying states, actions, transitions, goals, costs, constraints, and other fundamental elements required for problem solving.
• Select and apply appropriate search, optimization, and learning algorithms to solve different types of problems.
• Analyze the advantages, limitations, and suitability of different algorithms and problem-solving strategies.
• Justify the main choices made in the design and implementation of Artificial Intelligence-based solutions.
• Independently develop and implement problem-solving strategies using Artificial Intelligence techniques.
• Critically evaluate the results obtained and compare alternative solutions according to performance and quality criteria.

• Introduction to Artificial Intelligence
  1.1. Fundamental concepts and evolution of Artificial Intelligence
  1.2. Main paradigms and approaches of Artificial Intelligence
  1.3. Areas and applications of Artificial Intelligence
• Intelligent and Rational Agents
  2.1. Concepts of agent and environment
  2.2. Rationality and performance measures
  2.3. Types and architectures of intelligent agents
  2.4. Reactive, state-based, goal-based, and utility-based agents
• Problem Formulation and Problem Solving
  3.1. Problem representation
  3.2. States, actions, transitions, goals, costs, and constraints
  3.3. State space and solution space
• State-Space Search Algorithms
  4.1. Uninformed search strategies
  4.2. Heuristic search strategies
  4.3. Evaluation and comparison of search strategies
• Solution-Space Search Algorithms
  5.1. Local search methods
  5.2. Optimization methods
  5.3. Evolutionary computation algorithms
• Applications of Artificial Intelligence in Industry
  6.1. Production planning and scheduling
  6.2. Industrial process optimization
  6.3. Predictive maintenance
  6.4. Quality control and inspection
  6.5. Logistics, transportation, and resource management
  6.6. Intelligent decision-support systems
• Introduction to Machine Learning
  7.1. Fundamental concepts of Machine Learning
  7.2. Supervised and unsupervised learning
  7.3. Simple classification and prediction models
  7.4. Introduction to artificial neural networks
  7.5. Applications of Machine Learning in industrial contexts

Assessment in this course unit follows a continuous/periodic assessment model and consists of two components:

• Theoretical Component - 14 out of 20 marks, assessed through two tests taken during the semester or, alternatively, through a final examination.

• Practical Component - 6 out of 20 marks, assessed through two practical assignments:

  • Practical Assignment 1 - Rational Agents and Problem Solving: 3 marks;

  • Practical Assignment 2 - Local Search and Evolutionary Computation: 3 marks.

Tests and examinations are closed-book.

Practical assignments are carried out individually and may only be submitted once, on the dates specified in the respective assignment briefs.

Assignments must be submitted through the Moodle platform.

The grade obtained in the practical component is valid for all assessment periods, including the regular assessment period, resit period, special assessment period, and any other assessment periods available to students with special status.

Attendance is mandatory, and students must attend at least 75% of the classes in order to pass the course unit.

NAO