Hours:
16 hours (4 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

This PhD course focuses on Web search and discusses the challenges in the three main areas of Web search: i) crawling, ii) indexing, and iii) query processing. The course introduces each area by discussing the state of the art in the field and by presenting the open research questions. The emphasis of the course is on query processing, an area where machine learning provides an important contribution to advance the state of art. After an introduction of the different query processing techniques, the course i) introduces supervised techniques explicitly focused to target the ranking problem, ii) discusses several efficiency/effectiveness trade-offs in query processing and iii) analyse several related optimization techniques. The course will also provide an overview of the query processing techniques employing deep neural networks. Two hands-on sessions will cover indexing and query processing of public Web collections.

Course Contents in brief:

1. Modern Web Search (4 hours)
   1. The web: history, peculiarities and the importance of the search.
   2. Anatomy of a modern Web search engine: crawling, indexing, query processing.
   3. Crawling: definition and application. Architecture of a modern crawler.
   4. Challenges in crawling the Web
2. Fast Indexes for Web search (4 hours)
   1. Data structures for indexing Web documents
   2. Modern techniques for efficient text retrieval
   3. Data structures for efficient k-NN search and retrieval over learned representations
   4. Challenges in indexing the Web
   5. Hands On: Indexing and basic query processing on a public Web collection
3. Machine learning in modern query processors (8 hours)
   1. Machine learning approaches for IR: Learning to Rank
   2. Efficiency/Effectiveness Trade-offs, Cascading Architectures
   3. Neural information retrieval and the role of pre-trained large language models
   4. Dense/Sparse retrieval
   5. Hands On: Learning to Rank and Deep Neural Networks for efficient Web search

Schedule:

1. 18/06/2024: 9:00 - 13:00
2. 19/06/2024: 9:00 - 13:00
3. 20/06/2024: 9:00 - 13:00
4. 21/06/2024: 9:00 - 13:00

Hours:
16 hours (4 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

The conception and development of a Spacecraft is a complex and rather long process requiring the effort of many people and often also a big financial investment. The latter being true especially for scientific Spacecraft, which are one-of-a-kind units with typically challenging technological developments needed to accomplish the goals they have been built for.

Spacecraft need to survive two critical phases: the launch and the journey and operation in harsh environment. In both cases, servicing the Spacecraft is not possible, therefore confidence in its survival is required before launch and thus testing.

This course focuses on Spacecraft testing, covering all the different methods, with emphasis on the two main test categories: mechanical testing and thermal testing. An overview of the sensors used is also given, including an outlook to the future, covering methodologies and technologies that might soon become day to day tools during environmental testing.

Course Contents in brief:

1. Module 1: Why we test? Overview of needs and environmental test activities
2. Module 2: Mechanical Testing – Vibration testing, different types and goals
3. Module 3: Thermal Testing – Close look to thermal testing and thermal facilities
4. Module 4: Sensors used for testing and needs for the future (e.g. Virtual Twin)

Schedule:

1. Day 1: 10/6/2024 - 4 hours PM (13:00-17:00)
2. Day 2: 11/6/2024 - 4 hours AM (9:00-13:00)
3. Day 3: 12/6/2024 - 4 hours AM (9:00-13:00)
4. Day 4: 13/6/2024 - 4 hours AM (9:00-13:00)
5. Day 5: 14/6/2024 - AM exam

Hours:
8 hours (2 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

This 8-hour course will cover a comprehensive exploration of machine learning techniques applied to medical imaging, leading towards the early detection of melanoma. The course will cater to a diverse audience, from beginners to intermediate practitioners in machine learning, and will blend theoretical foundations with practical applications.

The course will begin with an overview of the iToBoS EU project, emphasizing the significant role of machine learning in enhancing the accuracy and efficiency of melanoma detection. This will set the stage for the technical content that follows, and highlighting the real-world impact of these technologies in healthcare.

Participants will then be introduced to the foundational concepts of machine learning, starting with linear regression, and their application in predictive analytics. The course will then transition into logistic regression, focusing on classification problems relevant to medical diagnostics.

As the course progresses, it will delve into neural networks, using logistic regression as a stepping stone to understanding neural networks. The progression from logistic regression to neural networks is a natural evolution from simpler linear models to more complex, non-linear models capable of handling a wider range of data and problem types in machine learning.

The course will culminate in an in-depth exploration of Convolutional Neural Networks (CNNs), vital for image recognition and analysis tasks in melanoma detection. Attendees will learn about the architecture of CNNs, their layers and filters, and the application of CNNs to dermoscopic images for melanoma detection.

Furthermore, the course will address the challenge of working with unbalanced datasets in medical imaging. It will introduce specialized evaluation metrics such as Precision, Recall, F1 Score, AUC-ROC, and the Precision-Recall Curve, which are essential for accurately assessing model performance in scenarios where traditional metrics may be misleading.

By the end of this course, participants will have gained a solid understanding of key machine learning concepts and their application in the medical field, particularly in melanoma detection.

Course Contents in brief:

Lecture 1: Learning Models (2 hours)

1. Overview of the iToBoS project
   • Early detection of melanoma
2. Supervised vs Unsupervised Learning
   • Understanding the structure of the data
   • Differentiating between regression and classification.
3. Predictive modeling: Linear Regression
   • Basic concept.
4. Modeling the hypothesis function
   • Formulating the hypothesis function.
   • Understanding linear regression parameters.
5. Cost Function and Gradient Descent
   • Mean Squared Error cost function
   • Implementing gradient descent.

Lecture 2: Logistic Regression (1 hour)

1. A linear classifier
   • Understanding logistic regression for classification.
   • Binary classification examples.
2. Hypothesis Representation and Decision Boundary
   • Sigmoid function and interpreting logistic regression output.
   • Concept of decision boundary in logistic regression.
3. Cost Function in Logistic Regression
   • Analyzing the cost function.
   • Convex optimization in classification.

Lecture 3: Basics of Neural Networks (2 hours)

1. Introduction to Neural Networks
   • Definition and motivation for using neural networks.
   • Overview of non-linear classification.
2. Perceptrons as Linear Classifiers
   • Introduction to the perceptron model.
   • Understanding how perceptrons work in linear classification.
3. From Linear to Non-Linear Boundaries
   • Limitations of linear classifiers like perceptrons.
   • The need for non-linear classification in complex problems.
4. Introduction to Neural Network Architecture
   • Basic components: neurons, weights, biases.
   • Understanding neuron layers and structure.
5. Learning with Neural Networks
   • Concept of feedforward and backpropagation.
   • Overview of the training process using examples.
6. Intuition on Backpropagation
   • Detailed explanation of backpropagation.
   • The significance of training neural networks.

Lecture 4: Advanced Concepts and Convolutional Neural Networks (2 hours)

1. Deep Dive into Backpropagation
   • In-depth analysis of the backpropagation process.
   • Computational graphs and gradient computation.
2. Understanding Convolutional Neural Networks (CNNs)
   • Distinction between Fully-Connected Networks (FCNs) and CNNs.
   • The architecture and operation of CNNs.
3. Components of CNNs
   • Convolutional layers, activation functions, and pooling.
   • Importance of ReLU, softmax functions, and dropout layers.
4. Training and Optimizing CNNs
   • Forward and backward propagation in CNNs.
   • Strategies to address vanishing/exploding gradients.
5. Application of CNNs in Image Recognition
   • How CNNs achieve feature extraction in image classification.
   • Practical examples of CNNs in image recognition.

Lecture 5: Case study for the detection of melanoma (1 hour)

1. Dealing with medical data
   • Bias in the data
2. Challenges of Unbalanced Datasets
   • Understanding the nature of unbalanced datasets in medical imaging.
   • Introduction to Precision, Recall, F1 Score, AUC.
3. Combining different data sources
   • Demographic information
   • Clinical history
   • Genomics
4. Importance of Accurate and Clinically Relevant Models
   • Relevant metrics in medical imaging: Sensitivity and Specificity

Schedule:

1. May 27, 2024: 14:00-17:00 (Lectures 1 and 2)
2. May 28, 2024: 14:00-16:00 (Lecture 3)
3. May 30, 2024: 14:00-16:00 (Lecture 4)
4. May 31, 2024: 14:00-15:00 (Lecture 5)

Hours:
16 hours (4 credits)

Room:

Aula Riunioni del Piano 6 del Dipartimento di Ingegneria dell’Informazione, Largo Lucio Lazzarino 1, Pisa

To register to the course, click here

Short Abstract:

Humans and other animals outperform traditional robots in terms of reliability and efficiency, largely due to their unique physical characteristics. They have elastic tendons, ligaments, and muscles, which allow robust interaction with the environment and dynamic task execution. In contrast, classic robots are often stiff and heavy. Accordingly, researchers in robotics have shifted from focusing on rigidity to incorporating lightweight and compliant structures. Drawing from natural examples, robots now include elastic and soft components, leading to the development of flexible joint and link robots and articulated and continuum soft robots. These latter types are composed entirely of deformable elements, making them akin to invertebrate animals. This proliferation of new robotic designs creates the challenge of developing effective control strategies for managing a nonlinear mechanical system with numerous degrees of freedom (DOFs) and a high degree of underactuation. This course will introduce this control challenge, especially focusing on continuum soft robots, review established findings, highlight recent advancements, and explore open issues in the field.

Course Contents in brief:

1. Introduction (2h)
2. Refreshers on model-based control and control of rigid robots (2h)
3. Modeling of continuum soft robots (4h)
4. Shape control (4h)
5. Task space control (2h)
6. Other control challenges in CSRs’ control (2h)

Schedule:

Il corso è stato riprogrammato a ottobre con le seguenti date e orari:

1. 07/10/2024: 9:00-10:30; breack;11:00-12:30;     14:00-15:30; breack; 16:00-17:15
2. 08/10/2024: 9:00-10:30; breack;11:00-12:30;     14:00-15:30; breack; 16:00-17:15
3. 09/10/2024: 9:00-10:30; breack;11:00-12:30;     14:00-15:30; 

Hours:
20 hours (5 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

Geometric Deep Learning (GDL) is an emerging interdisciplinary field that combines the power of deep neural networks with geometric and topological methods to tackle complex learning tasks.
This Ph.D. course offers a comprehensive exploration of the principles, theories, and applications of GDL, providing students with the necessary tools to understand and contribute to this exciting research area.
The course begins with an introduction to the fundamentals of deep learning, including neural networks, optimization, and training algorithms. From there, we delve into the world of geometry, exploring foundational concepts such as graphs and geometric transformations. We investigate how geometry can be seamlessly integrated with deep learning techniques to enhance the understanding and analysis of structured data.
In addition to theoretical knowledge, the course incorporates hands-on practical sessions where students gain experience with implementing and applying GDL algorithms/models with the Pytorch and the Pytorch Geometric libraries.
By the end of this course, students will have a comprehensive understanding of geometric deep learning, enabling them to apply cutting-edge techniques to a wide range of domains. They will be equipped with the knowledge and skills necessary to conduct original research, contribute to the development of novel algorithms, and advance the frontiers of GDL.

Course Contents in brief:

1. Logistic Regression for Neural Networks
2. A brief introduction to Neural Networks
3. Graph Convolutional Networks
4. Message-Passing Neural Networks
5. Pytorch/Pytorch Geometric

Schedule:

1. Day 1 – 9-13. Intro to Logistic Regression for neural networks and to the Pytorch Python library.
2. Day 2 – 9-13. A Brief Introduction to Neural and Convolutional Networks. Applications and implementation.
3. Day 3 – 9-13. Combining Feature-based and Geometric-based information in neural models. Graph Neural Networks.
4. Day 4 – 9-13. Extending the Graph Neural Networks: Message-Passing Neural Networks.
5. Day 5 – 9-13. Applications of MPNN in engineering.

Hours:
20 hours (5 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

Machine Learning has assumed a dominant role in the design of intelligent systems and their various application domains. The objective of this course is to present current trends in the development of Machine Learning, identify challenges and discuss ways of addressing them.

Course Contents in brief:

Introductory comments

The key agenda of Machine Learning. Main concepts. Deployment of Machine Learning and fundamental quests. Challenges of Machine Learning: credibility (confidence), interpretability and explainability, privacy.

Granular Computing: a primer

Concepts, motivation, examples. Design of information granules, rule-based architectures: symbolic- subsymbolic perspective. Learning schemes. 

Credibility of ML architectures and their results

Motivation. Granular embedding and Gaussian Process augmentation. Mechanisms of active learning.

Interpretability and explainability

Processes of interpretability and explainability. Inductive and deductive reasoning. Counterfactual reasoning.  Local linear models. Shapley value. 

Privacy in ML: a case of federated learning

Motivating factors behind federated learning: coping with data islands, average and gradient federated learning, Federated learning-based rule design, granular assessment and performance analysis.

Schedule:

1. 19/4/2024, 13.30-18.30 
2. 22/4/2024, 8.30-13.30
3. 23/4/2024, 8.30-13.30 
4. 24/4/2024, 8.30-13.30 

Hours:
16 hours (4 credits)

Room:

Aula Riunioni del Piano 6 del Dipartimento di Ingegneria dell’Informazione, Largo Lucio Lazzarino 1, Pisa

To register to the course, click here

Short Abstract:

This course provides participants with the chance to explore quantum computing, communication protocols, and the latest advancements in the field of the quantum internet. Following an initial overview of quantum computing and the current range of computing platforms, participants will delve into algorithms and measurement techniques. The instructor will cover the fundamentals of communication, teleportation, and entanglement. Practical exercises and hands-on experiences are integrated throughout the course.

Course Contents in brief:

1. Circuit Quantum Electrodynamics (QED) and the Jaynes-Cummings Hamiltonian
2. Characterization of superconducting qubits, measure coherence and population on multi-level quantum systems
3. Analysis of key figures of merit for quantum communication, such as entanglement rate and entanglement generation probability
4. Analyze, execute, and debug Python code as they implement various quantum circuits using various software packages.

Schedule:

1. 09/07/2024: 10:00-13:00; 14:30-17:00, Aula Riunioni del Piano 6 - Largo Lucio Lazzarino 
2. 10/07/2024: 10:00-13:00; 14:30-17:00, Aula Riunioni del Piano 6 - Largo Lucio Lazzarino 
3. 11/07/2024: 10:00-13:00; 14:30-16:30, Aula Riunioni del Piano 6 - Largo Lucio Lazzarino 

Hours:
20 hours (5 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

In recent years it has become clear that nonlinear dynamical systems have an extremely important role in modeling phenomena of different origin (physical, biomedical, economic, etc). I intend to present the basic notions of the theory of nonlinear dynamical systems and their applications to real world time series. With the help of simple examples, it will be possible to understand the notions of “chaotic” systems and of “complexity”.

Course Contents in brief:

1. Introduction to discrete dynamical systems and ergodic theory.
2. Measures of “chaos”: notions of entropy, Lyapunov exponents, algorithmic information content.
3. Random perturbations of dynamical systems.

Schedule:

1. 13/05/2024: 9:00 - 13:00
2. 14/05/2024: 9:00 - 13:00
3. 15/05/2024: 14:00 - 18:00
4. 16/05/2024: 14:00 - 18:00
5. 17/05/2024: 9:00 - 13:00

Hours:
20 hours (5 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

This short course has three aims:

1. Providing a critical-historical account of long enduring effort to develop humanized machines.
2. Providing insight from large multidisciplinary stance of theories and implementations
3. Describing the author’s own effort to develop a sentient humanoid

Course Contents in brief:

- HISTORIES OF AUTOMATA  ( OPTIONAL )

          From Hero to Voucanson

          From Wiener to Walters

          From Ashby to McCarthy

          From McCarthy to Brooks

- The rise of Humanoids

            Androids, gynoids and cyborg

            Just like a human? The Uncanny Valley controversy

            Internal robotics and machine monism

- Robotics meets neuroscience

              Cognitive neuroscience and the rise of affectivism

              Affect and emotions

              Edelman neural darwinism

              Damasio theory of consciousness

              Craig on interoception and inner time

-Robotics meets phenomenology

                Time of physics and subjective time: Einstein and Bergson

                From Husserl to Varela

                The embodied mind and the neurophenomenology program

                 Emotions, feelings and the self

                 A syncretic model of time-consciousness

-Towards sentient robots

                  A situated,embodied ,enactive machine

                   Body awareness and the protoself

                    Time-consciousness and the minimal self : core cosciousness

                    The autobiografical self and the extended consciousness

-Do we need conscious robots? Ethics and societal needs

Schedule:

1. 19/3/2023 - 9:00-12:00
2. 20/3/2023 - 9:00-12:00
3. 21/3/2023 - 9:00-13:00
4. 25/3/2023 - 9:00-12:00
5. 26/3/2023 - 9:00-13:00
6. 27/3/2023 - 9:00-12:00

Hours:
16 hours (4 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

This graduate course is focused on the study of online optimization and reinforcement learning applied to wireless and mobile mobile networks (IoT, 5G). Such networks may vary rapidly over time, potentially in an unpredictable and non-stochastic way because of ad-hoc user connectivity and behavior, and, hence traditional methods based on static (classic) or stochastic optimization and game theory are no longer suited. Instead, online optimization can be exploited to derive efficient algorithms, with theoretical guarantees in terms of no regret, that solve optimization problems where no assumptions can be made on the underlying temporal dynamics governing the network and, hence, the objective function to be optimized.

Course Contents in brief:

1. Iterative online process based on strictly causal feedback information
2. Regret measure: definition, no-regret property, intuition, links with static (classic) and stochastic optimal solutions
3. Link with multi-armed bandits from reinforcement learning (UCB, epsilon-greedy, EXP3 algorithms)
4. First-order online algorithms: online gradient descent, online mirror descent, and their theoretical guarantees in terms of no-regret and regret decay rates
5. Applications in wireless communications: beam-alignment in mmWave networks, energy-efficient NOMA power allocation, resource optimisation in IoT networks…
6. Beyond wireless: online metric learning for multimedia indexing, online matrix completion for movie ratings, universal filtering, etc.
7. Tradeoff between performance (regret decay) vs. required feedback information:
   • Feedback reduction: imperfect gradient feedback (stochastic gradient estimation), zeroth order methods (gradient estimation based on one value of the objective function)
   • Second order online descent methods
8. Lab practice (4 hours): implement and evaluate several multi-armed bandit algorithms to solve an outage minimization problem in a two-user adaptive NOMA system without any CSIT/CDIT, but relying solely on a single bit of feedback

Schedule:

1. 27/05/2024 – 8h-12h
2. 28/05/2024 – 8h-12h
3. 29/05/2024 – 8h-12h
4. 30/05/2024 – 8h-12h