![]() |
Educational guide | |||||||||||||||||||||||||||||||||||||||
IDENTIFYING DATA | 2024_25 | |||||||||||||||||||||||||||||||||||||||
Subject | ADVANCED MACHINE LEARNING | Code | 00717019 | |||||||||||||||||||||||||||||||||||||
Study programme |
|
|||||||||||||||||||||||||||||||||||||||
Descriptors | Credit. | Type | Year | Period | ||||||||||||||||||||||||||||||||||||
6 | Compulsory | Second | Second |
|||||||||||||||||||||||||||||||||||||
Language |
|
|||||||||||||||||||||||||||||||||||||||
Prerequisites | ||||||||||||||||||||||||||||||||||||||||
Department | ING.ELECTR.DE SIST. Y AUTOMATI |
|||||||||||||||||||||||||||||||||||||||
Coordinador |
|
ealeg@unileon.es efidf@unileon.es |
||||||||||||||||||||||||||||||||||||||
Lecturers |
|
|||||||||||||||||||||||||||||||||||||||
Web | http:// | |||||||||||||||||||||||||||||||||||||||
General description | ||||||||||||||||||||||||||||||||||||||||
Tribunales de Revisión |
|
|||||||||||||||||||||||||||||||||||||||
Competencias |
Code | |
A18964 | |
A18968 | |
A18973 | |
B5802 | |
B5803 | |
B5806 | |
B5807 | |
C2 | CMECES2 That students know how to apply their knowledge to their work or vocation in a professional manner and possess the skills that are usually demonstrated through the development and defense of arguments and the resolution of problems within their area of study. |
C4 | CMECES4 That students can transmit information, ideas, problems and solutions to both a specialised and non-specialised audience |
Learning aims |
Competences | |||
Knows advanced intelligent systems and deep learning techniques. | A18964 A18968 A18973 |
B5806 |
|
Applies, analyses and selects advanced intelligent systems and deep learning procedures to design solutions to problems. | A18964 A18968 A18973 |
B5802 B5806 |
|
Designs solutions based on artificial neural networks. | A18973 |
B5802 |
|
Interprets results with initiative, creativity and critical reasoning. | B5802 B5807 |
||
Transmits knowledge, information, ideas and solutions. | B5803 |
C2 C4 |
Contents |
Topic | Sub-topic |
BLOCK I. THEORY | 1. Introduction to Deep Learning 2. Review of neural network concepts. 3. Regularisation for Deep Learning 4. Optimisation for training deep models 5. Convolutional Networks 6. Sequence Modelling: Recurrent and Recursive Networks 7. Attention Models and Transformers 8. Autoencoders 9. Reinforcement Learning 10. Deep generative models |
BLOCK II. LABORATORY | Using the Python programming language, students will work on different problems to reinforce the concepts seen in theory and also to facilitate the realisation of the course project. |
Planning |
Methodologies :: Tests | |||||||||
Class hours | Hours outside the classroom | Total hours | |||||||
Presentations / expositions | 4 | 6 | 10 | ||||||
Practicals using information and communication technologies (ICTs) in computer rooms | 22 | 18 | 40 | ||||||
Problem solving, classroom exercises | 4 | 2 | 6 | ||||||
0 | 10 | 10 | |||||||
Assignments | 8 | 36 | 44 | ||||||
Lecture | 20 | 20 | 40 | ||||||
(*)The information in the planning table is for guidance only and does not take into account the heterogeneity of the students. |
Methodologies |
Description | |
Presentations / expositions | The students will explain during the theoretical classes the project carried out using a presentation previously prepared following the indications provided by the lecturer. |
Practicals using information and communication technologies (ICTs) in computer rooms | The practices of the course will be carried out using Python (3.X). Available in the F3 laboratory, although it is recommended that each student also install it on their personal computer. It is recommended to use Python 3.X from the installation of the Anaconda environment (https://www.continuum.io/downloads). Preferably the Spyder IDE, included in this installation, will be used. |
Problem solving, classroom exercises | Exercises are carried out using a variety of media such as a tablet, blackboard and also slides and examples using Excel. Exercises can also be left on Agora, with solutions, most of the time also with examples solved in Excel. |
The student will be offered online courses, typically on MOOC platforms, of very short duration to reinforce and complete the formation received. | |
Assignments | Students will carry out a project during the course that will be based on the theoretical contents explained in the classes and will be able to use some of the practical work carried out in the laboratory. |
Lecture | Theoretical sessions in the classroom using slides. Presentations or documents corresponding to the materials of each lesson may also be left on Agora. Lessons may be accompanied by videos related to the concepts presented, some recorded by the instructors and others from internet resources that the instructors consider particularly appropriate. Some lessons will be accompanied by a test with questions, which may be both theoretical and practical, and which could be assessed. It is foreseen that courses from the Datacamp platform, or similar courses, will be used to reinforce some of the lessons taught, with some courses being optional and others compulsory. |
Personalized attention |
|
|
Assessment |
Description | Qualification | ||
Practicals using information and communication technologies (ICTs) in computer rooms | The delivery of the labs indicated as compulsory, as well as the complementary activities indicated as compulsory, will be valued as 10% of the total grade. |
10 | |
Lecture | The contents taught in theory classes, both theoretical concepts and exercises, will be assessed by written exams, which will account for 50% of the course mark. | 50 | |
Some labs and other activities indicated will be voluntary and will be worth 10% of the course grade. It is possible to pass the course without submitting any of the voluntary activities, although the grade corresponding to the completion of these activities will not be obtained. |
10 | ||
Assignments | The student will carry out a project during the course, which will be assessed by means of the deliveries and presentations made and the corresponding written validation exams. The project must be completed and passed in order to pass the course. This project will be worth 30% of the course mark and will be assessed by means of presentations and written validation exams. Half of this 30%, 15%, corresponds to the mark for the presentations and the other 15% to the mark for the validation exam(s). | 30 | |
Other comments and second call | |||
The contents taught in theory classes, both theoretical concepts and exercises, will be assessed by written exams, which will account for 50% of the course mark. The student will carry out a project during the course, which will be assessed by means of the deliveries and presentations made and the corresponding written validation exams. The project must be completed and passed in order to pass the course. This project will be worth 30% of the course mark and will be assessed by means of presentations and written validation exams. Half of this 30%, 15%, corresponds to the mark for the presentations and the other 15% to the mark for the validation exam(s). The delivery of the labs indicated as compulsory, as well as the complementary activities indicated as compulsory, will be valued as 10% of the total grade. Some labs and other activities indicated will be voluntary and will be worth 10% of the course grade. It is possible to pass the course without submitting any of the voluntary activities, although the grade corresponding to the completion of these activities will not be obtained. In order to pass the course it is necessary to obtain 50% of the maximum mark for all the compulsory activities, which are all those indicated above except for the activities indicated as voluntary.
|
Sources of information |
Access to Recommended Bibliography in the Catalog ULE |
Basic |
![]() ![]() ![]() ![]() ![]() ![]() |
Ian Goodfellow, Yoshua Bengio and Aaron Courville. Deep Learning (Adaptive Computation and Machine Learning series) Illustrated Edition. The MIT Press, 2016. Online version: https://www.deeplearningbook.org/ Aston Zhang, Zachary C. Lipton, Mu Li and Alexander J. Smola. Dive into Deep Learning. Cambridge Univerity Press. 2023 Francois Chollet. Deep learning with Python, Manning, 2nd edition, 2021. Andrew S. Glassner. Deep Learning: A visual Approach. No Starch Press. 2021 Lewis Tunstall, Leandro von Werra and Thomas Wolf.Natural Language Processing with transformers: Building Language Applications with Hugging Face. O’Relly. 2022. Mohamed Elgendy. Deep Learning for Vision Systems. Manning. 2020 |
|
Complementary |
![]() ![]() |
Lewis Tunstall, Leandro von Werra and Thomas Wolf.Natural Language Processing with transformers: Building Language Applications with Hugging Face. O’Relly. 2022. Mohamed Elgendy. Deep Learning for Vision Systems. Manning. 2020 |
Recommendations |
Subjects that it is recommended to have taken before | |||||||
|