Traditional classes

Dear Students,

Decide when you would like to deliver your presentation.
It should be from the area of your studies and it should be 10-13 minutes long. Do remember to prepare a PPT presentation.

21 June 2019  Room 120

1.00 p.m.

1.20 p.m.

1.40 p.m.

2.00 p.m.

2.20 p.m.

24 June 2019 Room 231

5.00 p.m.

5.20 p.m.

5.40 p.m.

6.00 p.m.

6.20 p.m.

6.40 p.m.

7.00 p.m.

25 June 2019 Room 231

5.00 p.m.

5.20 p.m.

5.40 p.m.

6.00 p.m.

6.20 p.m.

6.40 p.m.

7.00 p.m.

26 June 2019 Room 231

5.00 p.m.

5.20 p.m.

5.40 p.m.

6.00 p.m.

6.20 p.m.

6.40 p.m.

7.00 p.m.

Week 7 [03-09.06.2019] Can you tell which face is real?

Hi!
Recently, there was hot news about a paper entitled "A Style-Based Generator Architecture for Generative Adversarial Networks". The new architecture leads to an automatically learned, unsupervised separation of high-level attributes (e.g., pose and identity when trained on human faces) and stochastic variation in the generated images (e.g., freckles, hair), and it enables intuitive, scale-specific control of the synthesis.

That research introduced a new, highly varied and high-quality dataset of artificial human faces.

The first approach to fake-face generation task was put together by Ian Goodfellow, the director of the machine learning at Apple’s Special Projects Group and a leader in the field when he proposed a new framework: Generative Adversarial Networks. In less than five years, all of that changed. Today’s AI-generated faces are full-colour, detailed images.

1. Have you ever heard about artificially generated images which contain realistic human faces? Can you propose some possible real-world applications of such algorithms?

You may have seen a website named ThisPersonDoesNotExist.com doing the rounds, which uses AI to generate startlingly realistic fake faces. There is also a website called WhichFaceIsReal.com, which lets you test your ability to distinguish AI-generated fakes from the genuine article.

2. Could you spot/are you aware of what kind of imperfection those generated faces have?

3. Do you consider any public security or privacy issues regarding those methods?

Articles and sources:

https://arxiv.org/abs/1406.2661
https://arxiv.org/pdf/1812.04948.pdf
https://www.youtube.com/watch?v=-cOYwZ2XcAc
https://www.theverge.com/2019/3/3/18244984/ai-generated-fake-which-face-is-real-test-stylegan
https://medium.com/@kcimc/how-to-recognize-fake-ai-generated-images-4d1f6f9a2842

Week 7 [03-09.06.2019] Modern medium for presenting research

Reading through a scientific paper is often arduous. As authors, we usually want to present our ideas as fast as possible to meet the deadline for a conference. As a consequence, we may introduce a research dept and transfer all the labor to readers. They'll have to struggle to understand our work.

To address the problem scientists mainly from Google and OpenAI started in 2017 a new journal called Distill. The journal is focused on clear and easy to understand presentation of ideas.

This is the article about the research dept: https://distill.pub/2017/research-debt/
And the journal itself: https://distill.pub/

Suggested questions to discuss:
1. Do you think that it's the authors of research papers responsibility to present their ideas in an easy to understand manner? Or doing it accurately and precisely is enough?
2. Are visualizations or interactive presentations any helpful for explaining ideas in your field of research? Or maybe other means of communication work better?
3.  Have you recently seen any work that helped you to understand some concept better?
I like this article about Attention Mechanism if you're interested: https://towardsdatascience.com/attn-illustrated-attention-5ec4ad276ee3

Week 7 [03-09.06.2019] Automated identification of media bias in news articles

I want to sharre with you an article with overview of news media bias detection methods. News media bias was well studied in social sciences and now become an interest of computer scientists. There are plenty of text mining, computer vision and other types of algorithms that may be used to detect media bias. I want to ask you the questions below.

https://link.springer.com/article/10.1007/s00799-018-0261-y

1. What is media bias? What types of media bias do you know?
2. Which type of media bias has the gratest impact on the reader?
3. Do you think it is possible to create automatic tools to detect news media bias?

Week 6 [20-26.05.2019] Something about music & medicine & technology

Dear All,

I'd like to share with my new interests:

https://positivepsychologyprogram.com/music-therapy-clinical/#dementia

The article indicates, how the music can influence on us, our health & diseases.

All of us know the music can help to get out of bad mood and feel better, sometimes even very fast. But this is not such issue.
I think the sounds (not only melodies) composed together in a specific way, can treat us from serious diseases. Some scientists also believe in hypotheses like that and lead specific researches on various detailed fields, like cancer, Alzheimer, Autism
And also, of course, many psychiatric cases use music therapy, like depression, schizophrenia etc.
But the most surprising for me were the effects achieved by scientists working with cancer cells, because there were many experiments proving decreasing or even liquidating them thanks to the sounds with specific frequency. Many other interesting issues I've met till now, I've read for example, that the sound of steps can help people after stroke with their legs' rehabilitation.

But coming to my questions:

1. Have you ever met with someone, who used such kind of healing (it may be either patient, or a doctor)?
2. Have you ever heard about technical tools enabling such therapies? If not, can you imagine such tool? What kind of?
3. Can you imagine that computer can support such treatments? How?
4. What was the most surprising for you in this article?
5. Do you consider this kind of methods as serious ones or just "magic" for naive people? Why?


Have a nice day,
Marta

Week 6 [20-26.05.2019] 3D printing of organs and tissues

Recently I came across a few articles about 3D printing of organs and tissues.

Sample articles:

https://mymodernmet.com/3d-printed-heart/

https://3dprintingindustry.com/news/japanese-researchers-use-scaffold-free-3d-bioprinted-tissues-for-regenerative-medicine-150838/ 

   The greatest success so far has been achieved by scientists from Tel Aviv University. They were the first to print an exact replica of the heart with blood vessels and chambers. The heart was printed from human cells and specific biological materials (substances consisting of sugars and proteins). The printed heart has the size of a rabbit's heart. His cells may shrink, but they must learn to pump so that they can cooperate.

   On the other hand, researchers from the University of Nagasaki and Saga are studying the method of printing 3D structures without scaffolding to create tissues for transplantation. Methods using scaffolds are widely known in medicine, but problems such as immunogenicity, scaffold degradation, and immune response may occur. Treatments based on scaffolding are also limited by materials that can be used to create scaffoldings. Thanks to the MCS (multicellular spheroids) method, in which the cells have the possibility of joining one another individually, the scientists were able to create the structures of esophageal tissue and then transplant them in rats. After surgery, none of the rats showed any complications.

   Biomaterials used in 3D printing of organs should have the same biochemical, mechanical and topographic properties as the patient's own tissues. By using biomaterials from patients, scientists eliminate the risk of implant rejection. This is extremely important because about 50% of transplants are rejected within 10-12 years of transplantation.

   3D printing in biomedicine creates great opportunities. Thanks to it, the waiting time for transplant can be shortened many times and the percentage of transplants rejected by the organism may be lowered.

1. What other organs or implants can be obtained by 3D printing method?

2. In the next 10-20 years, do you think that 3D bioprinters will be present in most hospitals, and most patients will have access to printed organs?

3. What technology could potentially replace the 3D printing of organs or be available faster?

4. Do you see any other advantages or disadvantages of the ability to implant 3D printed organs?

Week 6 [20-26.05.2019] The weather prediction with AI

Hello everybody,
today I would like to present two articles on the use of artificial intelligence in predicting the weather. After reading the articles, please answer the following questions:

1. Do you think that artificial intelligence will be used on a large scale in predicting the weather?
2. Will the weather predicted using artificial intelligence (numerical prediction of weather) be more accurate than the weather forecasted by applying weather data to weather maps?
3. Why do you need very efficient computers to predict the weather, or maybe you can speed up calculations in a different way?
4. Where can weather prediction be used?
Articles: 1. https://interestingengineering.com/ai-might-be-the-future-for-weather-forecasting2. https://medium.com/syncedreview/ai-local-rainfall-forecasting-using-weather-radar-maps-d4aa041343a