Subject: Applied Generative Artificial Intelligence
Scientific area:
Computer Science
Contact hours:
56 Hours
Number of ECTS:
4,5 ECTS
Language:
Portuguese
General Objectives:
1 - Provide students with a basic understanding of the fundamentals of artificial intelligence (AI), its evolution to modern models.
2 - Empower students to use Large Language Models (LLMs), integrating generative artificial intelligence into software platforms.
3 - Teach practical techniques for developing generative applications using frameworks.
4 - Explore use cases relevant to the tourism sector.
5 - Discuss the ethical, privacy, and security implications of using generative AI.
2 - Empower students to use Large Language Models (LLMs), integrating generative artificial intelligence into software platforms.
3 - Teach practical techniques for developing generative applications using frameworks.
4 - Explore use cases relevant to the tourism sector.
5 - Discuss the ethical, privacy, and security implications of using generative AI.
Contents / Syllabus:
1 - Introduction to Artificial Intelligence (AI)
1.1 - Definition of AI
1.2 - History and evolution of AI (symbolic AI, machine learning, and generative AI)
2 - Fundamentals of Machine Learning
2.1 - Concepts of supervised, unsupervised, and reinforcement learning
2.2 - Use of foundation models
3 - Generative Models
3.1 - What is generative AI
3.2 - Types of models (text, image, and multimodal)
3.3 - Large Language Models (LLMs)
4 - Integration of LLMs via API
4.1 - Authentication, calls, limits and costs
4.2 - Good prompt engineering practices
4.3 - Local model execution
5 - Development of Generative Applications
5.1 - Chatbots and conversational agents
5.2 - Augmented Recovery by Generation (RAG)
5.3 - Use of frameworks (LangChain): configuration, prompt pipeline, memory and agents
5.4 - Deployment of a web and mobile platform
6 - Evaluation and Monitoring
6.1 - Evaluation metrics for generated outputs
6.2 - Feedback and continuous adjustment
7 - Ethics, Privacy and Security
7.1 - Bias in generative models
7.2 - Data privacy and GDPR
7.3 - Regulation and responsible use
7.4 - Social impact and implications for tourism (personalization and privacy of customer data)
1.1 - Definition of AI
1.2 - History and evolution of AI (symbolic AI, machine learning, and generative AI)
2 - Fundamentals of Machine Learning
2.1 - Concepts of supervised, unsupervised, and reinforcement learning
2.2 - Use of foundation models
3 - Generative Models
3.1 - What is generative AI
3.2 - Types of models (text, image, and multimodal)
3.3 - Large Language Models (LLMs)
4 - Integration of LLMs via API
4.1 - Authentication, calls, limits and costs
4.2 - Good prompt engineering practices
4.3 - Local model execution
5 - Development of Generative Applications
5.1 - Chatbots and conversational agents
5.2 - Augmented Recovery by Generation (RAG)
5.3 - Use of frameworks (LangChain): configuration, prompt pipeline, memory and agents
5.4 - Deployment of a web and mobile platform
6 - Evaluation and Monitoring
6.1 - Evaluation metrics for generated outputs
6.2 - Feedback and continuous adjustment
7 - Ethics, Privacy and Security
7.1 - Bias in generative models
7.2 - Data privacy and GDPR
7.3 - Regulation and responsible use
7.4 - Social impact and implications for tourism (personalization and privacy of customer data)
Bibliography / Information Sources:
I. Goodfellow, Y. Bengio e A. Courville , 2016 , Deep Learning , MIT Press
J. Alammar e M. Grootendorst , 2024 , Hands-On Large Language Models: Language Understanding and Generation , O?Reilly Media
J. Berryman e A. Ziegler , 2024 , Prompt Engineering for LLMs: The Art and Science of Building Large Language Model-Based Applications , O?Reilly Media
J. Phoenix e M. Taylor , 2024 , Prompt Engineering for Generative AI: Future-Proof Inputs for Reliable AI Outputs , O?Reilly Media
B. Singh , 2024 , Building Applications with Large Language Models: Techniques, Implementation, and Applications , Springer
J. Alammar e M. Grootendorst , 2024 , Hands-On Large Language Models: Language Understanding and Generation , O?Reilly Media
J. Berryman e A. Ziegler , 2024 , Prompt Engineering for LLMs: The Art and Science of Building Large Language Model-Based Applications , O?Reilly Media
J. Phoenix e M. Taylor , 2024 , Prompt Engineering for Generative AI: Future-Proof Inputs for Reliable AI Outputs , O?Reilly Media
B. Singh , 2024 , Building Applications with Large Language Models: Techniques, Implementation, and Applications , Springer
Assessment Methods and Criteria:
Classification Type: Quantitativa (0-20)
Evaluation Methodology:
This course combines theoretical and laboratory classes, favoring a practical approach to the use of generative AI models. Activities include developing applications with LLMs, API integration, prompt engineering, use of frameworks, and analysis of use cases in tourism. Assessment integrates a practical project (50%), focused on developing a generative AI-based application, and a written exam (50%), designed to evaluate the understanding of fundamental concepts and ethical and technological implications.
Evaluation Methodology:
This course combines theoretical and laboratory classes, favoring a practical approach to the use of generative AI models. Activities include developing applications with LLMs, API integration, prompt engineering, use of frameworks, and analysis of use cases in tourism. Assessment integrates a practical project (50%), focused on developing a generative AI-based application, and a written exam (50%), designed to evaluate the understanding of fundamental concepts and ethical and technological implications.


