AIMS AND CONTENT

LEARNING OUTCOMES

The course aims to introduce the student to the fundamental machine learning techniques used in automated driving, with a focus on the perception aspect. The student will be stimulated to develop analytical and design skills also through the production of an exam work on a real-world case study.

AIMS AND LEARNING OUTCOMES

The main objective of the course is for the student to gain a broad knowledge of state-of-the-art deep learning techniques for automated driving. For each topic covered, the student will have the opportunity to learn the theoretical foundations and application code. Examples are offered in order to verify the acquisition of knowledge and stimulate its application. The course examples will use the Python language and the libraries sk-learn, sk-time, Keras/Tensorflow.

The project that will be agreed upon for the examination is aimed at stimulating and verifying the student's design and implementation skills, as well as the operational verification in the field of a significant part of the acquired knowledge.

The learning outcomes concern the realisation of the aforementioned learning objectives, also through the implementation of a project. At the end of the course the student will be able to analyse and design state-of-the-art deep learning solutions for context perception and prediction of its evolution.

TEACHING METHODS

Lectures face-to-face, with use of slides, and examples/exercises carried out on the PC, mainly using the sk-learn, sk-time, Keras/Tensorflow libraries, in python language. Student reception. Proposal, implementation and discussion of a project.

SYLLABUS/CONTENT

• Introduction to machine learning for automated driving
• Training & pre-training of multilayer perceptrons
• Optimizers
• Regularization
• CNN architectures
• CNN training
• Object detection
• Semantic segmentation / panoptic
• Yolop
• Graphnet
• Timeseries classification and prediction
• LSTM, 1D convolution
• Transformers
• Autoencoders
• Generative adversarial networks
• Explainable machine learning
• Unsupervised machine learning (clustering and anomaly detection)

RECOMMENDED READING/BIBLIOGRAPHY

A. Geron, Hands-On Machine Learning With Scikit-Learn and Tensorflow: Concepts, Tools, and Techniques to Build Intelligent, O’ Reilly

I. Goodfellow, Y. Bengio and A. Courville, Deep Learning, The MIT Press

Documentation of the used libraries

Lecture notes and other material suggested by the lecturer during the course

Non-attending students may contact the lecturer to agree on the best arrangements.

Students with disabilities can contact the lecturer to agree on the best arrangements.

TEACHERS AND EXAM BOARD

Exam Board

FRANCESCO BELLOTTI (President)

LUCA LAZZARONI

MARCO RAGGIO

RICCARDO BERTA (President Substitute)

LESSONS

LESSONS START

https://easyacademy.unige.it  

Class schedule

The timetable for this course is available here: Portale EasyAcademy

EXAMS

EXAM DESCRIPTION

Project work on an example of machine learning application for automated driving (knowledge, understanding, analysis, judgement, application, creation in relation to the topics covered in the lecture)

ASSESSMENT METHODS

Evaluation will take place at the various stages of project preparation: definition interviews, design/implementation of the solution, final discussion of a paper describing the work done.

The lecturer will also keep in mind the student's participation during the course.

Exam schedule