Understanding understanding + LSTM based Autoencoders
Details
I am happy to announce the next DL meetup, the event will be on 17th of September 2019. There will be 2 presentations, and we are going to be at the same location as last time: Millenaris Startup Campus (we are supported by IVSZ yet AGAIN).
- presentation: Antal Jakovac (professor at Eötvös Loránd University)
https://www.linkedin.com/in/antal-jakovac-93493084/
The title of his presentation is “Understanding understanding”, he will try to show what is “understanding”, how intelligence works and why do we have to do so many trials/experiments (in ML) before we succeed. You will also suggest how the ML community should modify their AI strategies to get more reliable models.
- Presentation: There are two presenters here: Nora Szaloczy (Data Science Team Leader at Spicy Analytics/Senior research executive)
https://www.linkedin.com/in/szaloczy-nora/
Valer Kaszas (Machine learning researcher and data scientist)
https://www.linkedin.com/in/valerkaszas/
They are going to talk about a real-life application in the health industry, they are searching for clusters in huge amounts of patient data, they will present their business case, and the solution also in theory and in practice with some source code! I’d like to spoiler here a little bit, they are using LSTM based Autoencoders and with some mixture of K-Means.
For the first presentation there is no need to read anything, I think. For the second one (LSTM based Autoencoders) you should read the following papers:
https://www.sciencedirect.com/science/article/pii/S1532046417300710
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.91.9082&rep=rep1&type=pdf
Hopefully, you will have time to go through these.
Location: Millenaris Startup Campus
https://www.millenaris-startupcampus.hu/
Budapest, Kis Rókus u., 1024, D Building
Networking starts at 18:00, the presentations will be started at 18:30.
We will have pizza and beverages as always.
The event is supported by IVSZ. ( http://ivsz.hu/ )
The event will be recorded on video.
Understanding understanding + LSTM based Autoencoders