Deep Learning
Faculteit | Science and Engineering |
Jaar | 2021/22 |
Vakcode | WMAI017-05 |
Vaknaam | Deep Learning |
Niveau(s) | master |
Voertaal | Engels |
Periode | semester II a |
ECTS | 5 |
Rooster | rooster.rug.nl |
Uitgebreide vaknaam | Deep Learning | ||||||||||||
Leerdoelen | This course aims to introduce the student to the field of Deep Learning; by the end of it the student will be familiar with: - The basics of supervised, unsupervised and reinforcement learning and how one can tackle problems of all three machine learning branches with neural networks. - The optimization methods underlying the training process of neural networks. - Different types of neural architectures ranging from feedforward neural networks and convolutional networks to recurrent neural networks and transformers. - The main research challenges currently characterizing the field of deep learning. |
||||||||||||
Omschrijving | Most of today's popular AI-based techniques are based on artificial neural networks, a family of machine learning algorithms that were originally introduced with the aim of mimicking the biological processes underlying the human brain. While most of the developments coming from today's deep learning community are not motivated by this research quest anymore, their resulting machine learning models have nevertheless achieved very impressive results. Neural networks can be used to tackle numerous machine learning problems, ranging from the recognition and classification of millions of natural images to mastering complicated board games such as Go and chess. Throughout this course, we will see what makes neural networks such powerful models, how one can train this family of techniques successfully, and what the future of deep learning currently looks like. | ||||||||||||
Uren per week | |||||||||||||
Onderwijsvorm | Hoorcollege (LC), Practisch werk (PRC) | ||||||||||||
Toetsvorm |
Opdracht (AST), Schriftelijk tentamen (WE)
(The final grade for the course will be based on two practical assignments and a final written exam. The assignments will account for 40% of the final grade, whereas the exam will count for 60%. Students need to score at least 5.0 on the written exam to pass the course. The final weighted grade should be at least 5.50.) |
||||||||||||
Vaksoort | master | ||||||||||||
Coördinator | M. Sabatelli, PhD. | ||||||||||||
Docent(en) | M. Sabatelli, PhD. ,Dr. M.A. Valdenegro Toro | ||||||||||||
Verplichte literatuur |
|
||||||||||||
Entreevoorwaarden | Mandatory: No prior knowledge is assumed. Please note that the student is expected to have a relevant BSc degree. Advised: Prior knowledge about Linear Algebra and Calculus is assumed. Because the course is intended for AI students, participants are supposed to have this prior knowledge. |
||||||||||||
Opmerkingen | This course unit has a capacity limit. More information about capacity-limit courses can be found here. This course has an intended limit of 100 participants. If there are more AI MSc students than 100, they are all permitted to take the course. The course unit prepares students to do their graduation project if they choose to do it in deep learning. |
||||||||||||
Opgenomen in |
|