Machine Learning 2
Code: 104871 ECTS Credits: 6| Degree | Type | Year | Semester |
|---|---|---|---|
| 2503852 Applied Statistics | OB | 3 | 2 |
Contact
- Name:
- Antonio Lozano Bagen
- Email:
- antonio.lozano.bagen@uab.cat
Teaching groups languages
You can check it through this link. To consult the language you will need to enter the CODE of the subject. Please note that this information is provisional until 30 November 2023.
Teachers
- Roger Borràs Amoraga
Prerequisites
The first year subjects, in addition to Numerical Methods and Optimization and Machine Learning 1.
Objectives and Contextualisation
To learn at theoretical and practical level the potentialities of deep learning for structured and also unstructured data.
Competences
- Analyse data using statistical methods and techniques, working with data of different types.
- Correctly use a wide range of statistical software and programming languages, choosing the best one for each analysis, and adapting it to new necessities.
- Critically and rigorously assess one's own work as well as that of others.
- Make efficient use of the literature and digital resources to obtain information.
- Select and apply the most suitable procedures for statistical modelling and analysis of complex data.
- Select statistical models or techniques for application in studies and real-world problems, and know the tools for validating them.
- Select the sources and techniques for acquiring and managing data for statistical processing purposes.
- Students must be capable of applying their knowledge to their work or vocation in a professional way and they should have building arguments and problem resolution skills within their area of study.
- Students must be capable of collecting and interpreting relevant data (usually within their area of study) in order to make statements that reflect social, scientific or ethical relevant issues.
- Students must be capable of communicating information, ideas, problems and solutions to both specialised and non-specialised audiences.
- Students must develop the necessary learning skills to undertake further training with a high degree of autonomy.
- Summarise and discover behaviour patterns in data exploration.
- Use quality criteria to critically assess the work done.
- Work cooperatively in a multidisciplinary context, respecting the roles of the different members of the team.
Learning Outcomes
- Analyse data using an automatic learning methodology.
- Critically assess the work done on the basis of quality criteria.
- Describe the advantages and disadvantages of algorithmic methods compared to the conventional methods of statistical inference.
- Develop a study based on multivariate methodologies and/or data mining to solve a problem in the context of an experimental situation.
- Discover individuals' behaviours and typologies through data-mining techniques.
- Identify the statistical assumptions associated with each advanced procedure.
- Identify, use and interpret the criteria for evaluating compliance with the requisites for applying each advanced procedure.
- Implement programmes in languages suitable for data mining.
- Make effective use of references and electronic resources to obtain information.
- Obtain and manage complex databases for subsequent analysis.
- Reappraise one's own ideas and those of others through rigorous, critical reflection.
- Students must be capable of applying their knowledge to their work or vocation in a professional way and they should have building arguments and problem resolution skills within their area of study.
- Students must be capable of collecting and interpreting relevant data (usually within their area of study) in order to make statements that reflect social, scientific or ethical relevant issues.
- Students must be capable of communicating information, ideas, problems and solutions to both specialised and non-specialised audiences.
- Students must develop the necessary learning skills to undertake further training with a high degree of autonomy.
- Use data mining methods to validate and compare possible models.
- Use summary graphs of multivariate or more complex data.
- Work cooperatively in a multidisciplinary context, accepting and respecting the roles of the different team members.
Content
Topic 1: Fully connected neural networks.
Topic 2: Convolutional neural networks.
Topic 3: Recurrent neural networks.
Topic 4: Other types of neural networks.
Methodology
Teaching will combine classroom lessons by teachers and practical work for students with a computer.
In all aspects of teaching/learning activities, the best efforts will be made by teachers and students to avoid language and situations that can be interpreted as sexist.
To achieve continuous improvement in this subject, everyone should collaborate in highlighting them.
Annotation: Within the schedule set by the centre or degree programme, 15 minutes of one class will be reserved for students to evaluate their lecturers and their courses or modules through questionnaires.
Activities
| Title | Hours | ECTS | Learning Outcomes |
|---|---|---|---|
| Type: Directed | |||
| Theoretical and practical classes | 50 | 2 | |
| Type: Autonomous | |||
| Personal study of the subject | 46 | 1.84 | |
| Practices | 30 | 1.2 |
Assessment
Continuous grading
The grading for the course will be done in two parts: the theory part, NT, and the practice part, NP. The final grade for the course will be N = 0.5*NT + 0.5*NP..
The grading fot the theory part will be based in two exams: a partial examn, NEP, and a final exam, NEF. The final grade for the theory part will be NT = max(NEF, 0.3*NEP + 0.7*NEF).
The grading for the practice part will have two parts: una continuous evaluation part, NPC, and one final project part, NPT. The final grade for the practice part will be NP = 0.5*NPC + 0.5*NPT.
On the day of the second-chance exam only the grade for the theory part will be updated. If a student goes to the second-chance examn then the theory grade, NT, will be the grade for the second-chance exam.
In order for an activity to be taken into account in the final grade, the activity grade has to be a minimum of 3,5.
The student who has submitted works for at least 50% of the subject will be considered evaluable. Otherwise, it will appear in the record as non-evaluable.
The grades of the second-chance exam will not be taken into account for the allocation of Honor Mentions.
Single grading
The grading for a student who chooses to be evaluated with the single grading modality will be N = 0.5*NEF+ 0.5*NPT.
Assessment Activities
| Title | Weighting | Hours | ECTS | Learning Outcomes |
|---|---|---|---|---|
| Exam | 50% | 4 | 0.16 | 1, 11, 2, 5, 3, 17, 6, 7, 8, 10, 15, 14, 12, 13, 18, 9, 16 |
| Task | 50% | 20 | 0.8 | 1, 5, 17, 8, 10, 4, 15, 14, 12, 13, 9, 16 |
Bibliography
- Geron, A. (2019) Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow (O'Reilly)
- Goodfellow, I. et al (2016) Deep Learning (MIT Press)
- Chollet, F. (2017) Deep Learning with Python (Manning)
Software
Python