Simone Scardapane

DATE Wednesday 11th of April 2018

LOCATION Rome, Polo Didattico - Piazza Oderico da Pordenone 3

This hands-on workshop gives a practical introduction to deep learning in Python. We show how to move from standard machine learning to deep neural networks using different levels of abstractions in TensorFlow, with a focus on the entire pipeline process (from data preprocessing to model evaluation). Take a look at the Special Package: “1 Workshop + Conference Ticket”: we offer 15% DISCOUNT on the total price!

Take a look at the Special Package: “1 Workshop + Conference Ticket”: we offer 15% DISCOUNT on the total price!
Click here to know how to obtain these discounts.

LANGUAGE
Italian

LEVEL
Beginner

DURATION
The workshop is full-day (8 hours) from 9:00 to 18:00, with one hour lunch break.

CHECK IN 8:30 – 9:00

PRICES
Every 8 hour workshop ticket is fixed: – to 160 € until the 2nd of February; – to 190 € until the 3rd of March; – to 220 € until the 29th of March; – to 250 € until the 10th of April at 5 pm Take a look at the Special Package: “1 Workshop + Conference Ticket”: we offer 15% DISCOUNT on the total price!

Take a look at the Special Package: “1 Workshop + Conference Ticket”: we offer 15% DISCOUNT on the total price!
Click here to know how to obtain these discounts.

SIMONE SCARDAPANE Simone Scardapane is a post-doc fellow at Sapienza University (Rome). His research is focused on machine learning, with an emphasis on deep learning, distributed environments, and applications in the audio field. He is co-founder of the Italian Association for Machine Learning, co-organizer of the Machine Learning Meetup in Rome, and a Google Developer Expert for Machine Learning. Before his PhD, he obtained a B.Sc. in Computer Engineering in 2009, and an M.Sc. in Artificial Intelligence and Robotics in 2011. He is an active member of several organizations, including the IEEE Computational Intelligence Society, the International Neural Networks Society, and the AI*IA.

ABSTRACT
The workshop will provide an overview to using machine learning and deep learning algorithms in Python. After recalling the basic concepts of supervised learning (e.g., optimization, overfitting), we will show how to implement two complete ML workflows (from preprocessing to model tuning) on realistic use-cases. After describing sklearn in the morning, we will show limitations of standard ML techniques when dealing with high-dimensional data, and explain how to design advanced deep learning models using TensorFlow. Everything will be implemented with live coding on interactive Jupyter environments.

TABLE OF CONTENTS
1- The workshop will begin by recalling the basic concepts of supervised learning and numerical computation in Python. 2- A complete workflow for classification using scikit-learn will be discussed. 3- We will show a complete use case of deep learning in TensorFlow, by explaining in-depth its mechanics and how it differs from sklearn. 4- Additionally, we will comment on advanced deep learning architectures (e.g., recurrent neural networks).

TRAINING OBJECTIVES
At the end of the workshop, participants will be able to design standard machine learning models with sklearn, and complex deep learning models in TensorFlow for their applications, in different tasks such as automatic prediction, classification, and high-dimensional data regression. Additionally, they will understand the difference between classical machine learning and deep learning.

WHO THE WORKSHOP IS DEDICATED TO?
The workshop is open to everyone interested in understanding the fundamentals of ML and deep learning, with an emphasis on Python libraries. The workshop will introduce both theoretical and practical concepts, together with their concrete implementation.

PREREQUIREMENTS
No prior exposure to machine learning or to numerical computation in Python is requested, as all the necessary concepts will be introduced when necessary. A basic knowledge of linear algebra and/or optimization is enough to navigate throughout the entire workshop.

HARDWARE AND SOFTWARE REQUIREMENTS
A standard laptop is enough to replicate all code provided in the workshop. In order to have a working installation of Python with all the required tools, participants can install the Anaconda distribution (https://www.continuum.io/downloads) or a similar full-stack solution. For the purposes of the workshop, a standard TensorFlow installation (even without GPU support) will be sufficient, see https://www.tensorflow.org/install/.

WARNING
Seats are limited.
The workshop will be held only if the minimun number of attendees is reached.

Take a look at the Special Package: “1 Workshop + Conference Ticket”: we offer 15% DISCOUNT on the total price!
Click here to know how to obtain these discounts.