In the winter semester 2018/2019, Dr. Muhammad Zeshan Afzal will give the “Very Deep Learning” lecture at the University of Kaiserslautern. The lecture will be organized with the OLAT system. Please access the course via this link. The password will be given in the lecture. Please hand in your first assignment there. You are encouraged to work in groups. However, every student should upload a solution file.
We just finished a great event in Berlin. The Ovation Summer Academy (OSA) 2017 was much more successful then ever expected. In this combined research exchange & hackathon event, 30 highly motivated open-minded researchers, students, and members from industry met in the Backfabrik in Berlin and generated six proof of concepts for Conversational AI.
The MindGarage and Insiders Technologies GmbH are working on Conversational Intelligence (CI) to make machines understand Natural Language as well as humans do. CI includes Chat-Bots to core modules like Named Entity Recognition, Intent Classification, Visual Question Answering, Question Answering and Reasoning, etc.. We deal with all these challenging tasks in our everyday work. To give Conversational Intelligence a lateral perspective and to figure out fresh solutions for them, we are bringing together computer scientists from all around Europe to what we call the **Ovation Summer Academy** (OSA).
During the last week two new publications have been uploaded on arxiv. I am happy and a bit proud that the Vision of the MindGarage leads to such nice research contributions. I use this chance to thank the members and supporters of the MindGarage for their active participation! Especially, Andreas, Ayushman, John, and Shan have done a great job! See the project desciptions in Multilevel context representation and Text Conditioned Auxiliary Classifier Generative Adversarial Network in the projects section.
Good news: Our paper on sensor data transformation has been accepted at the International Joint Conference on Neural Networks (IJCNN 2017). See you in Alaska! The project description is in the projects section.
Today is the last lecture on Very Deep Learning before Christmas. For those who missed some of the lectures, you can always join the MindGarage in the OLAT E-learning Environment of the University of Kaiserslautern and get access to the lecture information and slides: MindGarage at OLAT. For the password, please send a mail to Marcus Liwicki.
After the official opening we celebrated the successful start with a glass of wine/juice and used the chance for some interesting discussions and for getting to know each other. Thanks to the kind support of the Insiders team, we had a nice atmosphere and good drinks. Also the chips I offered (various kinds of all brands one could get in the region of Kaiserslautern) were all eaten at the end of this celebration. I am happy that so many people (>50) came and that we had such interesting discussions about AI, possible projects and collaborations with other research groups, and the societal aspects of AI. Below you can find some impressions of the evening. Let us celebrate the MindGarage being “The place for OpenMinded Research”!
Today it was the time of celebration. We officially opened the MindGarage. We are very happy that the Vice-President of the University, Prof. Norbert Wehn gave a very nice welcome address where he sketched the history of the Mindgarage and AI in Kaiserslautern. Also, he underlined that the leadership team of the University of Kaiserslautern supports the MindGarage and they will find a location where we can stay on the long-term. Thank you!
After visionary talks of Marcus Liwicki and Werner Weiss (CEO of Insiders Technologies), two student members of the Mindgarage, John and Ayushman, presented their view of the MindGarage. They are happy that the MindGarge is a place to meet and interact, they are happy to have direct discussions with the experienced researchers Marcus and Shan, and that - last but not least - the MindGarage feels like a family. This has been said so nicely that I almost began to cry (see pictures below :) )
After the short opening speeches, there was a chance for networking and exchanging research ideas (subject of the next post). All in all it was a very successful evening and I am very thankful to everyone who came and especially who helped making this evening and the MindGarage such a success (see last slide as picture below).
Last week we have set-up the MindGarage. As a small token of appreciation, Marcus invited the volunteers to a dinner at the Saigon restaurant. It was a gorgeous dinner and we had nice discussions about what we are going to do in the MindGarage. Some impressions below.
By the end of October 2016 we received the computers and began setting them up as well as moving the last furniture into the MindGarage. Here you find some impressions of our activities!
Thanks for all the volunteers who helped and made everything possible!
Starting in autumn 2016 we formed a kind of independent “lab” at the University of Kaiserslautern which we call the MindGarage. The Ovation MindGarage is a place for students and researchers who really want to experience Very Deep Learning. It offers a set of High-End Powerful GPU-Computers for performing any experiments the students would like to conduct without bureaucratic constraints. The MindGarage is lead by Prof. Dr. Marcus Liwicki with the help of Dr. Muhammad Zeshan Afzal and a highly motivated team of Master students. The infrastructure is financially supported by Insiders Technologies GmbH, the rooms and administrative support is provided by the University of Kaiserslautern.
I am greatful for all support given and looking forward to a successful research/industry partnership leading to novel research perspectives and applications in Kaiserslautern.
Subscribe to this blog via RSS.
Seq2seq (2) Intro (2) Fun (3) Rnn (1) Event (9) Official (1) Lecture (1) Very deep learning (1) Footstep detection (1) Transfer learning (1) Publications (2) Ijcnn (1) Music information retrieval (1) Object recognition (1) Context representation (1) Arxiv (1) Image understanding (1) Image generation (1) Generative adversarial neural networks (1) Document classification (2) Deep learning architectures (2) Mindstorm (5) Segmentation (1) Documents (1) Deep learning (2) Gans (1) Adversarial (1) Teaching (1)