About DLS 2020


UCA Event 2020, from July 6-10


In response to the pandemic, this year the UCA DLS is replacing its "Deep Learning" Summer School with a fully virtual program. This program will be online and safe, free of charge, open to more attendees and participants will be able to attend online and at their own time.


Whether you are a researcher, an Engineer, an expert in "Deep Learning" or eager to learn more about those new capabilities essential for the future, this program is for you. It includes:


• A series of 4 online webinars "Deep Learning foundations "; this format allows you to attend at your convenience and learn at your pace;


• 5 "Expert Labs "moderated by subject matter experts; those will take place from 2-5pm every afternoon from July 6-10 

Webinar Series

“Deep Learning foundation” online recorded webinars  

The 4 modules below are dedicated to beginners who have heard about neural networks and machine learning without knowing exactly how it works.

 Each module ranges in duration for 40’’ to 50’’. 

  • Webinar 1: we will start by a general and brief overview of what is machine learning, followed by explaining the first model of artificial neuron and its difference with the biological model. 
  • Webinar 2: we will then present why putting this neuron in a network and how to train an artificial neural network. 
  • Webinar 3: we will explain why evolving towards deep neural networks and what specific optimization techniques and tips and tricks these models require to converge.
  • Webinar 4: we will then go through the different kind of deep neural networks, the different structures, to give you a global view of all possibilities of deep learning.

There is no specific requirement to follow this lecture, but if you have basics in vectors, and derivatives, it will be easier. 

Even if the access is free this year, you need to register to access the webinars

SPEAKER

precioso


Since 2011, Frederic Precioso is Professor at Université Côte d'Azur, lecturer at the École d'ingénieur Polytech'Nice Sophia, member of the SPARKS (Scalable and Pervasive softwARe and Knowledge Systems) team of the I3S UMR 7271 CNRS-UNS laboratory.
He is member of Maasai, a Joint Research Project between INRIA-CNRS-UCA.
His main research interests are: Machine learning, Deep Learning, for many application domain.
Since September 2018, he is Scientific and Program Officer for the National Research Agency (ANR) for State Investment Programmes Division and Digital Technology and Mathematics Dept.

Expert Labs

5 Deep Learning Expert labs are organized over the 6-10th July week, from 2-5 pm CET time
Each lab is designed around practical cases and covers an independent topic.
You can choose to follow one or all of them depending on your interest.
Those labs are moderated by 2-3 subject matter experts allowing individual interactions
To attend those Expert labs, you are expected to have an intermediate to advanced level on the topic. If you are a beginner, the only chance to possibly be able to follow the labs is to attend to all of them, and to already have strong basis in Python.
Space is limited and registration invitation only.

Lab 1 - July 6th 

The first lab will also be dedicated to beginners to visualize how (deep) neural networks work, how to build your first network, how to babysit it to make it converge, how to modify hyperparameters to converge faster, better, and with more confident results. We will start exploring structuring neural networks with Convolutional Neural Networks. We will also see how to initialize a new network from a pre-trained network built for another task or another domain with Transfer Learning.

Lab 2 - July 7th 

The second lab will be dedicated to better explore Convolutional Neural Networks (CNNs), their behavior, and how to adapt their internal structure to define explore different algorithms for Object Detection, Instance Segmentation and Semantic Segmentation on images.

Lab 3 - July 8th 

The third lab will focus on Recurrent Neural Networks (RNNs), how they can be used to analyze sequential data, and how combining them with attention mechanisms improve results in the context of sentiment analysis in texts.

Lab 4 - July 9th 

The fourth lab will be dedicated to audio data analysis and speech recognition. In this lab, we will experiment how deep learning works with audio signals; more specifically, we will learn how to build and train some efficient deep learning models to recognize speech by combining CNNs, RNNs, and Attention mechanisms.

Lab 5 - July 10th 

The fifth lab will be dedicated to Deep Reinforcement learning. This lab is meant to provide a first experience on using Deep Reinforcement Learning (DRL), for both synthetic and more realistic problems. While applications of RL are typically limited to discrete, low-dimensional constraints, recent advances in Deep RL (DQN for Atari 2600, AlphaGo, and more lately AlphaGo Zero) have demonstrated human-level or super-human performance in complex, high-dimensional spaces. 

To become a partner, please

Our community

Over past years, together we built a strong community of experts around Deep Learning Skills  competencies essential for the future….    

Over the past 3 years, more than 600 engineer researchers, PHD students coming from the Academic world or from the Business world shared ideas , experiences and great time together. 

They come from: 

Our community

And Many Others...