Politecnico di Milano
Funzioni disponibili

Risorse bibliografiche
Risorsa bibliografica obbligatoria
Risorsa bibliografica facoltativa
Scheda Riassuntiva
Anno Accademico 2018/2019
Scuola Scuola di Ingegneria Industriale e dell'Informazione
Insegnamento 052372 - NEUROENGINEERING [I.C.]
  • 052370 - NEUROENGINEERING [1]
Docente Cerveri Pietro
Cfu 5.00 Tipo insegnamento Modulo Di Corso Strutturato
Didattica innovativa L'insegnamento prevede  1.0  CFU erogati con Didattica Innovativa come segue:
  • Blended Learning & Flipped Classroom
  • Soft Skills

Corso di Studi Codice Piano di Studio preventivamente approvato Da (compreso) A (escluso) Insegnamento

Obiettivi dell'insegnamento

The goal of Neuroengineering part [1] is the study of engineering methods and technologies to further the ability to understand and augment brain function, and taking advantage from this knowledge to use computational models to address problems not only in the biomedical field but also in other applications. By means of a transdisciplinary approach that joins biologic and mathematical areas, the course aims to provide specific physiological insights about high-level human brain abilities such as reasoning, recognition and memorization, along with engineering tools to mimic such abilities. Computational models will be described and analyzed with emphasis on sub-symbolic computation with special focus on artificial neural networks and deep learning techniques to solve function representation, data modeling, pattern classification, and information storage and retrieval problems.

Risultati di apprendimento attesi

In the domain of the presented subjects, the students are expected to achieve:

- knowledge and understanding with a cross-disciplinary methodology about artificial neural networks together with the ability of selecting the appropriate models for the specific applications taking into consideration user, operational and technical requirements, ending with proper validation metrics to test the suitability and reliability of the selected model. (lectures) - DD 1

- ability to implement neuro-engineering solutions into specific biomedical research and clinical scenarios, starting from a proper understanding of artificial neural network methods. (lectures and practical exercises)- DD 2

- capability to judge the problem assumptions in the light of the obtained results (validation accuracy, degree of complexity, computational cost) and the scalability of the solution in different scenarios of applications. (lectures and discussion teaching workshops) – DD 3,5

- ability to effectively work in group, and communicate by public speaking. (discussion teaching workshops)- DD 4,5


Argomenti trattati

Neuroengineering part [1] will address three main topics:

 - ARTIFICIAL NEURAL NETWORK PRINCIPLES: this part will focus upon the formulation of the artificial neural networks (ANN) as a computational tool mediated from human brain structure and functions for classification, feature extraction and function approximation tasks. The students will be provided with tools to understand the link between reinforced learning and neural plasticity with mathematical models of the formal working neuron, the emerging properties of neural populations and the training mechanisms based on error backpropagation. Feed-forward and feed-back neural architectures will be described along with supervised and unsupervised learning approaches. The students will learn how to select the specific neural network typology for the particular problem to address, making practice about the use of ANN for complex function reconstruction and pattern classification in biomedical applications.

- DEEP LEARNING TECHNIQUES: this part will capitalize on ANN methodologies to extend the modeling ability of the feed-forward  networks in the representation of information by means of multiple layers of progressive encoding as the human brain does when processing and storing for example the sensorial information. Examples about the human visual system will help to understand how to deploy ANN-based computation models able to represent information in depth. Convolutional and autoencoder networks will be discussed as models for implementing deep learning strategies with specific exemplifications and real-life applications as gesture and emotion recognition from video images.  The students will make practice using advanced SW tools and devoted toolboxes. 

- NEURAL MEMORIES: this part will describe the main features of the human memory from both physiological and modeling points of view by focusing on episodic storage. Reinforcement learning will be revised under the paradigm of synaptic connection potentiation in the long-term memory. Recurrent networks will be described as computational tools to implement neural memories able to store and retrieve information. Procedures for pattern encoding and decoding will be presented in the paradigm of the Hebbian learning. The implementation of auto-associative memories will be described by means of binary network evolution, describing both parallel and sequential dynamics. The student will learn the criteria for optimal information storage, accuracy of pattern recall and network stability.  The students will make practice about pattern modeling, encoding and memory implementation using advanced SW tools.

The course is organized as follows:

  • Phase 1: The professor of Neuroengineering [1] will provide frontal classes and practical lessons to introduce the students to the contents of its part (about 35-40 Hours)
  • Phase 2: The professor of Neuroengineering [2] will provide frontal classes and seminars to introduce the students to the contents of its part. A step-by-step evaluation is additionally proposed at the end of every topic, their average can substitute the written exam of this part. (about 35-40 Hours)
  • Phase 3: Discussion teaching workshops (about 30 Hours) INNOVATIVE TEACHING CREDIT

All the students are organized into small groups (about 8 students), led by a tutor (usually a PhD student or a Post-doc). The two professors propose an even number of topics. Each students choses a topic (or if necessary, is assigned to a topic to assure roughly uniform numbers on each topic). All the students on one topic form a team (about 8-10 students), led by a tutor (usually a PhD student or a Post-doc).

Each team deeply studies the chosen research topic, starting from a couple of scientific papers proposed by the tutor.

Meeting 1: the team, facilitated by the tutor, discusses the state of the art and defines the goal of a project to improve what presented in the literature.

Meeting 2: Tailored on the goal of the project proposed by the first meeting, the tutor prepares a lab experience, in order to provide the group of the best available practical knowledge to solve the goal.

Autonomous team work:

Each group is then divided by the professor into subgroups of about 3 students, which from now on compete in a challenge on the same topic. All subgroups work independently to implement a proposed solution of the goal. Each subgroup can meet twice with the tutor.

Final presentation:

In the final week of the course, all subgroups present their solution and a board composed by all tutors and professors evaluated the proposed solutions and give a mark to each subgroup, considering multiple aspects such as: originality of the idea, feasibility, communication skills. Prizes will be awarded.


Basic knowledge of neurophysiology, numeric calculus and programming.

Modalità di valutazione

A written test is used to assess the acquired knowledge of the topics, the ability to analyse use cases scenarios, design proper solutions including assumptions, requirements, specifications and validations methods.

The final score is composed by the average of the two written tests of  Neuroengineering [1] and [2] (this latter can be substituted by the step-by-step tests). The score of the two written tests is scaled at a maximum of 27/30, then an additional 6/30 max is added depending on the evaluation of the Discussion teaching workshops.

If there are exceptional reason not to participate to the discussion teaching workshop, custom solutions will be defined in order to verify the achievement of specific objectives.

Risorsa bibliografica obbligatoriaCourse powerpoint slides BEEP
Risorsa bibliografica obbligatoriaCourse note compendium BEEP
Risorsa bibliografica obbligatoriaSolved exercise tests BEEP
Risorsa bibliografica obbligatoriaSource codes (matlab) for exercise laboratory BEEP
Risorsa bibliografica obbligatoriaWriteen exam tests with aswers BEEP

Forme didattiche
Tipo Forma Didattica Ore di attività svolte in aula
Ore di studio autonome
Laboratorio Informatico
Laboratorio Sperimentale
Laboratorio Di Progetto
Totale 50:00 75:00

Informazioni in lingua inglese a supporto dell'internazionalizzazione
Insegnamento erogato in lingua Inglese
Disponibilità di materiale didattico/slides in lingua inglese
Possibilità di sostenere l'esame in lingua inglese
Disponibilità di supporto didattico in lingua inglese
22/03/2019 Area Servizi ICT v. 1.4.11 / 1.4.11