EMERGING TECHNOLOGIES: AI, MACHINE LEARNING, BLOCKCHAIN, IOT, 5G, QUANTUM & EDGE COMPUTING

EMERGING TECHNOLOGIES: AI, MACHINE LEARNING, BLOCKCHAIN, IOT, 5G, QUANTUM & EDGE COMPUTING

Alessandro Chessa, Davide Carboni

Obiettivi formativi

The course aims at introducing students to relevant emerging technologies, providing a basic but in-depth conceptual understanding of Blockchain and decentralised ledgers, Machine Learning and Artificial Intelligence, Network Science with applications of IoT and 5G especially to Smart Cities and Territorial Networks, highlighting its potential applications in the social, economic and legal fields, with particular attention to the renewed sustainability challenges. Students are encouraged to apply theoretical notions in practical sessions in order to solve empirical problems through a hands-on approach with intuitive visual tools. They are also encouraged to envisage new projects involving these technologies and try to solve small real world cases of interest.

Prerequisiti

The only prerequisite is a genuine curiosity in understanding emergent technologies and to be passionate about learning the implications that these technologies have over society.

Contenuti Del Corso

This course is intended to transmit sophisticated technological concepts and its implications for society in a straightforward, motivating and easy-to-understand way, avoiding at the same time unnecessary technicalities and extreme simplifications. We will explore blockchain and decentralised ledgers, machine learning and artificial intelligence, 5G and IoT, with a particular attention to the impact that all these technologies have in shaping the society we live in. At the end of the twelve weeks our students will be able to understand when emergent technologies are bringing a real added value to a specific project, and when instead they are inserted just for fashion or for “hype”, i.e. without a real need. First module After a brief introduction to the main concepts of programming, which will be useful throughout the course, the first six weeks of the course will be devoted to an in-depth exploration of the vast blockchain universe. Blockchain was officially invented in 2009 by Satoshi Nakamoto (which is just a nom de plume, as his/her/their real name has always been a mystery), but its origins date back to the 80’s and 90’s of the last century, when a group of people started to imagine how cryptography could allow exchange of information and creation of trust among individuals in a decentralised and private way. Blockchain is more than meets the eye, and cryptocurrencies are only the tip of the iceberg. Therefore, in order to fully understand the potential of blockchain in the first module we will delve into the concepts of competition and cooperation, of decentralisation and consensus, we will explore examples from society and from nature, to understand how players of a game can be incentivised to cooperate together for a common good without the need of a central authority. We will also learn what it means to mine a block, what is a smart contract, we will talk about Bitcoin, Ethereum and other cryptocurrencies, tokens and NFts, but always with the aim of understanding why blockchain can play a pivotal role in shaping the future of our society. Second module The field of Machine Learning and Artificial Intelligence have a history dating back to the 1950s and in the seventh week of the course (the first of the second module), after a brief historical introduction, all the basic concepts of the methodology will be introduced, which differs greatly from traditional algorithm design. In the following week the four main methodological categories will be explored: regression, classification, clustering and dimensionality reduction. In the ninth week of the course we will explore a highly topical subject, Network Science, the currently most powerful paradigm for understanding the complex dynamics of interactions in social and economic phenomena. Networks are also a fundamental starting point to understand the shaping of Smart Cities supported by the IoT technologies. In the tenth we will lay down the theoretical foundations of Artificial Neural Networks that we will deepen in the next one talking about Deep Learning, the latest application frontier of AI in solving the most challenging problems, with performance similar to the capabilities of human thought. The last week will be dedicated to the discussion of various issues related to the impact of AI in our society, such as Explainable AI, AI Crime, ethical problems linked to these emerging technologies and finally to the creative possibilities of machines.

Testi Di Riferimento

First module: “Blockchain: Blueprint for a New Economy” by Melanie Swan (book) “The Tragedy of the Commons” by Garrett Hardin (paper) “The Evolution of Cooperation” by Robert Axelrod (book, optional) Second module: “On Intelligence” by Jeff Hawkins - Macmillan (book) “Superintelligence” by Nick Bostrom - Oxford (book) “Artificial Intelligence” by Alessandro Vitale - Egea (book in italian) “Computing Machinery and Intelligence” by Alan Turing - Mind (paper: https://www.csee.umbc.edu/courses/471/papers/turing.pdf, https://en.wikipedia.org/wiki/Computing_Machinery_and_Intelligence)

Metodologie Didattiche

On line/on site lectures live exercises with students Case studies with the direct involvement of students Problem based learning Peer education

Modalità di verifica dell'apprendimento

The student's knowledge will be assessed in two different ways: 1. One final project-oriented group assignment on a topic that will possible include elements of both modules and which will be proposed by the groups and have to be accepted by the teachers. 2. Two personal assignments in the form of written homework, in which the students will have to write a mini-paper for each of the two modules of the course (for an overall length of around 1.000 words per paper) on a specific topic assigned to each student by the teachers. The aim of the project-oriented group assignment is to give the students a way to test their capability in understanding the added value that emerging technologies can bring to a specific project. Ideally, the students should be able to point out why the projects they are working on for their assignment are only feasible with the aid of the proposed technology. The teachers will also value the capability of students to bring their own knowledge into the project, e.g., exploring the legal or the economic aspects deriving from the implementation of a given technology within the project. The project-oriented group assignment will count for 1/3 of the final grade. The aim of the personal assignment is to assess the capability of the student to explore in deeper detail a specific subject, starting from the knowledge received from the course. The form of a short mini-paper (approximately 1000 words) is chosen to stress the importance of expressing the concepts in a concise and understandable way (quality over quantity). The personal assignments will count for 2/3 (1/3 for each module) of the final grade. Finally, at the end of each week there will be an exchange of views between the students and the teachers on the topics discussed during the week, in the form of a class debate. The students will be asked to elaborate a very short informal assignment in written or oral form, either before the class, or immediately before the debate. These oral or written elaborations will be shared with the class as a primer for the debate. These intermediate assignments and the following debate will not be evaluated, as they are intended as a moment in which students can learn together via trial and error, through direct confrontation with the class and sharing of ideas.

Criteri per l’assegnazione dell’elaborato finale

For the personal assignment, the topic will be assigned by the teacher. For the group assignment, the groups will be randomly formed by the teachers and the subjects will be decided by the groups. The teacher can accept or modify the subject of the group depending on its relevance to the topics treated throughout the course.

Settimana 1

Online Overview of the course The birth of coding: Ada Lovelace and Charles Babbage What is a programming language (overview and simple coding examples) Coding examples: Cellular Automata, Tit for Tat The Bitcoin blockchain: an historical introduction Campus The birth of blockchain: bitgold, hashcash, the double spending problem and the concept of proof of work Hashing, blocks, mining, minting The Bitcoin blockchain and the Bitcoin Q&A session

Settimana 2

Online The emergence of Cooperation Decentralisation and Consensus The example of ants as a natural decentralised system The Byzantine attack Campus Distributed ledgers Permissioned VS permissionless blockchains Q&A session

Settimana 3

Online Ethereum Smart Contracts The Ethereum Virtual Machine and the Gas Campus Decentralised Applications (Dapps) Soft and Hard Forks Decentralised Autonomous Organisations (DAO) The DAO attack The “Blockchain is Law” motto and its implications Q&A session

Settimana 4

Online Alternatives to the Proof of Work: Proof of Stake, Proof of Space, Proof of Elapsed Time Sustainability and ecological impact: the issues of the proof of work protocol Campus Cryptocurrencies Is Bitcoin just a bubble? Privacy-by-design and zero-knowledge proof Q&A session

Settimana 5

Online ICOs and Whitepapers Practical applications: an overview of some blockchain-based projects Campus Blockchain and emergent Technologies: 5G, IoT, Machine Learning and AI Q&A session

Settimana 6

Online The tragedy of the Commons Tokens and NFts: overview and practical applications Campus The future of blockchain: is blockchain here to stay? How could blockchain change the world in the future? Blockchain and society Blockchain and ethics Q&A session

Settimana 7

AI, Machine Learning, Deep Learning: an historical introduction Supervised, Unsupervised learning Training, validate and testing Regularization: overfitting and underfitting Hyperparameter tuning Performance metrics Feature engineering

Settimana 8

Linear and Logistic Regression Classification Clustering Dimensionality Reduction

Settimana 9

Network Science and Complexity Centrality Measures, Motifs, Community Detection Recommender systems Urban/Territorial Networks, IoT and Smart Cities Knowledge Graph, Semantic Web and Ontologies Applications in sustainability and SDGs

Settimana 10

Artificial Neural Networks (ANN) Neuroscientific motivations and ANN Layouts Constructing and Training Neural Networks Optimization processes

Settimana 11

Deep Learning CNN, RNN, LSTM Reinforcement Learning Chatbots GPT3, Transfer Learning

Settimana 12

AI and Society AI Ethics XAI: Explainable AI Deep Fake and AI Crime Deep Dreams and Creative Machines