Deep Learning: Introduction, Architectures and Use Cases

Hello wonderful big data developers and enthusiasts. We hope this email finds everyone well!

We are happy to announce our fifth event for 2017 which will take place this Monday, December 11, 2017 (just 3 days away!) at The Cube Athens (venue has changed!). This time our agenda will be a bit more broad hosting two interesting talks about deep learning. By doing so, we hope we can attract a bigger audience, including but not limited to, data scientists and machine learning engineers. After all, we believe that big data and data science go hand in hand and thus we hope that both presentations will be beneficial for engineers and scientists alike! More about the talks bellow.

The event will start at 7:00 PM as usual, but this time in a difference place: The Cube Athens (Kleisovis 8, Athens[masked], Greece).

We are really looking forward to seeing you there and don’t forget to spread the word!

Adrianos ( | Euangelos ( | Stavros (


1st Talk:

Deep learning for content recommendation at high-scale

Taboola is the world’s leading content discovery platform. The challenge we face is selecting the most suitable content for each user in a given context, in less than a second and out of millions of available items. In this lecture we will discuss Taboola’s high-scale, deep learning solution for content recommendation and its real world challenges. We will then dive into the solution architecture which combines neural networks and matrix factorization concepts and discuss some of our key challenges.

Dr. Gil Chamiel ( is a Director of Data Science and Algorithm Engineering at Taboola ( Gil holds a PhD in Computer Science (AI) from the University of New South Wales, Australia in the area of personalization and preference elicitation. He is a Taboola veteran and has been working on Taboola’s core algorithmic engine for 7 years.

2nd Talk:

Deep learning for modeling visual and textual modalities: research and applications

In this talk we present an introduction of modeling visual and textual modalities using deep learning. We will talk about convolutional neural network (CNN) based architectures for object classification and image understanding, as well as recurrent neural networks (RNN) for text modeling and topic/sentiment classification. Finally, we will highlight indicative examples of applications for the case of recommender systems, e-commerce product categorization, image captioning and visual question answering.

Dr. Theodorakis Stavros ( is a co-founder and senior research engineer at DeepLab ( DeepLab develops machine learning solutions for real-word applications and production systems, while bridging the gap between the research and industry. Stavros holds a PhD in machine learning from the National Technical University of Athens, Greece, and has worked as a research assistant in EU-funded research projects and applied machine learning solutions to real-world applications.


7:00 – Socialize
7:25 – Welcome
7:30 – 1st Talk
8:15 – 2nd Talk
9:00 – Drinks & Snacks

We are always looking for speakers for our meetups. If you would like to give a talk please contact with Adrianos (, Euangelos (, or Stavros (