Tillämpad Deep Learning med Tensorflow

5 hp

Deep neural networks är den metod som vanligtvis används vid till exempel bildanalys i självkörande fordon och robotik. Denna kurs riktar sig till yrkesverksamma med lite erfarenhet inom området, till exempel en grundläggande kurs i Machine Learning, som vill lära sig mer om Deep Learning. Kursen ingår i kurspaketet RELIFE (hh.se/relife) där du som deltagare kan läsa hela kurspaketet eller enstaka kurser. Kursen är för yrkesverksamma och ges på distans på engelska. Anmälan är öppen så länge det finns möjlighet att bli antagen.

In English: The course is included in the course package RELIFE (hh.se/relife) where participants can take the entire course package or individual courses. The course is for professionals and is held online in English. Application is open as long as there is a possibility of admission. The courses qualify for credits and are free of charge for participants who are citizens of any EU or EEA country, or Switzerland, or are permanent residents in Sweden. More information can be found at antagning.se.


About the course Applied Deep Learning with Tensorflow
Deep neural networks have taken over as the preferred method in various fields, such as image analysis in autonomous driving and robotics. This course gives professionals with some experience in the area, for example those who have taken a basic Machine Learning course, an opportunity to learn or sharpen their theoretical and practical skills on deep learning methods while working remotely from home.
The course "Applied Deep Learning with Tensorflow" is split into two modules of 2.5 credits each:
1. Theory (8 x 45 min)
2. Practice (8 x 45 min)

The theoretical content covers basic principles of multi-layer perceptrons, spatio-temporal feature extraction with CNNs and RNNs, classification and regression of big data, and producing novel data samples using generative models. The practical sessions cover the basics of programming with tensorflow, image classification, and semantic segmentation using CNNs, future image frame prediction with RNNs and image generation with GANs.

Participants will be encouraged to bring their own data to solve problems they face in their own work. Students without related experience should first take the course Supervised Machine Learning. The course instructors have experience teaching a similar course in our master programme. Guest lecturers will be invited from the company Zenuity to ensure industry relevance. Halmstad University has several high performance computers dedicated to students to apply their deep learning experiments.

This course is for professionals with some experience in the area, for example those who have taken a basic Machine Learning course. The course is held online in English.

HT 2021 (Distans (Internet), Ortsoberoende, 50%)

Nivå:

Avancerad nivå

Anmälningskod:

13311

Behörighetskrav:

Högskoleingenjörsexamen eller Teknologie kandidatexamen samt Kontrollerad Machine Learning 7,5 hp.

Urvalsregler:

Högskolepoäng: 100%

Mer information om urvalsregler.

Startvecka:

Vecka: 35

Obligatoriska sammankomster:

0

Undervisningstid:

Blandade tider

Studieavgift:

För sökande med medborgarskap utanför EU/EES och Schweiz: Mer information om studieavgift

Undervisningsspråk:

Undervisningen bedrivs på engelska.

Visa utbildningsinfo

Sidansvarig

Dela