Omfattning: 5

Tidtabel: 09.09.2019 - 09.12.2019

Kontaktuppgifter till kursens personal (gäller denna kursomgång): 

Leo Kärkkäinen (leo.karkkainen at aalto.fi)

I am available in general at Aalto on Thursdays and Fridays.  I will be available for answering questions on Thursdays 11:00 - 11:30 at  F304, Otakaari 3.

Course assistants:

Christabella Irwanta (christabella.Irwanta at aalto.fi) and Ruslan Lagashkin ( lagashkinruslan at gmail.com)




Undervisningsperiod (är i kraft 01.08.2018-31.07.2020): 

I-II 2018-2019 (autumn)

Lärandemål (är i kraft 01.08.2018-31.07.2020): 

After the course, the student

  • can explain the fundamentals of neural networks function.
  • is able to construct, design, implement and run a deep learning based AI analysis project in medical domain.
  • has the skills to select and utilize and the tools for applying AI for the different use cases needed for medical applications.
  • is able to identify and develop the data sets that provide the best results.

Innehåll (är i kraft 01.08.2018-31.07.2020): 

The course will contain the basics in deep learning, introduce the student how machine learning is used in different medical domains, and what technologies are appropriate in different use cases. The work has mathematical and programming exercises,  and a project for designing and teaching a neural network for a medical use case.

 

Närmare beskrivning av kursens innehåll (gäller denna kursomgång): 

Lectures Mondays and Exercises on Fridays  8-10 am at TU6 (1199), TUAS, Maarintie 8, Exercises Fridays

  • Lecture 1:  What is currently possible with AI? Selection of project work topics.
  • Lecture 2: Introduction to probability. Basics and what is needed for the course.
  • Lecture 3: Machine learning. Unsupervised, supervised, and reinforcement learning
  • Lecture 4: Neural networks and medical data, legislation and ethics in use of AI. Data augmentation
  • Lecture 5: Training of networks, back-propagation, stochastic gradient descent (SGD). Models of data with spatial, (translation) symmetries. Convolutional  neural networks (CNNs). Medical imaging, object classification
  • Lecture 6: Models of data with temporal symmetries. Recurrent neural networks.

    Physiological signals, like ECG and BCG. Modeling the cardiovascular  system.
  • Lecture 7: Autoencoders, predicting the future, and de-noising signals. Generative adversarial networks (GANs)
  • Lecture 8: Natural Language Processing. Attention models. Generating text
  • Lecture 9: AI in embedded systems.  Compressing deep neural networks. HW for AI.
  • Lecture 10: Challenges in using AI in health. Adversarial examples. Differential privacy
  • Lecture 11: What could we expect of the future?
  • Lectures 12 - 13 and Exercise  13: Project work reports
  • Lecture 14: Exam

Metoder, arbetssätt och bedömningsgrunder (är i kraft 01.08.2018-31.07.2020): 

Project work, exercises, and final examination

Närmare information om bedömningsgrunderna och -metoderna och om hur den studerande kan ta del av bedömningen (gäller denna kursomgång): 


Exercises: 
Each week we will have some derivations and
questions related to the course content, and programming tasks. At the
exercise you have to mark the ones you have solved and be prepared to
present these.

At minimum finish 50% of the exercises.
20% of final score is based on amount of exercises completed in an acceptable level.

Project work
At the beginning of the course some project work
topics will be provided, but you can also suggest your own according to
your personal preference. You can work on this alone are in pairs.

At the end of the course, you will give a lecture and demonstration of your work at the lecture time.

Project work is 40% score

Final exam.
40% of the score.

Arbetsmängd (är i kraft 01.08.2018-31.07.2020): 

48 h contact teaching, 85 h independent studies

Studiematerial (är i kraft 01.08.2018-31.07.2020): 

See MyCourses

Förkunskaper (är i kraft 01.08.2018-31.07.2020): 

Python progamming and linear algebra. Mathematica and Matlab skills will be useful

Bedömningsskala (är i kraft 01.08.2018-31.07.2020): 

0-5

Anmälning (är i kraft 01.08.2018-31.07.2020): 

WebOodi

Tilläggsinformation (är i kraft 01.08.2018-31.07.2020): 

Language class: 3

Beskrivning

Anmälning och tillläggsinformation