Second EJD STIMULATE School on "Fundamentals of Data"

4 Feb 2019, 08:30 → 15 Feb 2019, 18:00 Europe/Rome

C412 (University of Ferrara - Department of Physics and Earth Sciences)

 

 

 

 

Week 1: Mon, Feb 4th – Fri, Feb 8th

	Mon 4th	Tue 5th	Wed 6th	Thu 7th	Fri 8th
08:30-09:00	Welcome
09:00–12:50	Fundamentals  of Deep Learning	Fundamentals  of Deep Learning	Data analytics	Data analytics	students project presentations
14:00–17:50	Exercises and Programming	Exercises and Programming	Exercises and Programming	Exercises and Programming

Week 2: Mon, Feb 11th – Fri, Feb 15th

	Mon 11th	Tue 12th	Wed 13th	Thu 14th	Fri 15th
09:00–12:50	Data storage and management	Data storage and management	Data storage and management	Applications and use cases	students project presentations
14:00–17:50	Exercises and Programming	Exercises and Programming	Exercises and Programming	Applications and use cases

Chair: Sebastiano Fabio Schifano

Data Format 1

Data Format 2

Data Format 3

Data Management 1

Data Management 2

Data Management 3

Data_Management_hands-on.tgz

Monday 4 February

08:30 → 09:00 Welcome

09:00 → 10:45 Fundamentals of Deep Learning

Fundamentals of Deep Learning for Computer Vision (Day 1)
In this hands-on course, you will learn the basics of deep learning by training and deploying neural networks. You will:
Implement common deep learning workflows such as Image Classification and Object Detection.
Experiment with data, training parameters, network structure, and other strategies to increase performance and capability.
Deploy your networks to start solving real-world problems.
On completion of this course, you will be able to start solving your own problems with deep learning.

10:45 → 11:05 Coffe Break

11:05 → 12:50 Fundamentals of Deep Learning

12:50 → 14:00 Lunch

14:00 → 15:45 Fundamentals of Deep Learning

15:45 → 16:05 Coffe Break

16:05 → 17:50 Fundamentals of Deep Learning

Tuesday 5 February

09:00 → 10:45 Fundamentals of Deep Learning

Fundamentals of Deep Learning for Multiple Data Types (day2)
This workshop explores how convolutional and recurrent neural networks can be combined to generate effective descriptions of content within images and video clips.
Learn how to train a network using TensorFlow and the Microsoft Common Objects in Context (COCO) dataset to generate captions from images and video by:
Implementing deep learning workflows like image segmentation and text generation
Comparing and contrasting data types, workflows, and frameworks
Combining computer vision and natural language processing
Upon completion, you’ll be able to solve deep learning problems that require multiple types of data inputs.

10:45 → 11:05 Coffe Break

11:05 → 12:50 Fundamentals of Deep Learning

Fundamentals of Deep Learning for Multiple Data Types (day2)
This workshop explores how convolutional and recurrent neural networks can be combined to generate effective descriptions of content within images and video clips.
Learn how to train a network using TensorFlow and the Microsoft Common Objects in Context (COCO) dataset to generate captions from images and video by:
Implementing deep learning workflows like image segmentation and text generation
Comparing and contrasting data types, workflows, and frameworks
Combining computer vision and natural language processing
Upon completion, you’ll be able to solve deep learning problems that require multiple types of data inputs.

12:50 → 14:00 Lunch

14:00 → 15:45 Fundamentals of Deep Learning

Fundamentals of Deep Learning for Multiple Data Types (day2)
This workshop explores how convolutional and recurrent neural networks can be combined to generate effective descriptions of content within images and video clips.
Learn how to train a network using TensorFlow and the Microsoft Common Objects in Context (COCO) dataset to generate captions from images and video by:
Implementing deep learning workflows like image segmentation and text generation
Comparing and contrasting data types, workflows, and frameworks
Combining computer vision and natural language processing
Upon completion, you’ll be able to solve deep learning problems that require multiple types of data inputs.

15:45 → 16:05 Coffe Break

16:05 → 17:50 Fundamentals of Deep Learning

Fundamentals of Deep Learning for Multiple Data Types (day2)
This workshop explores how convolutional and recurrent neural networks can be combined to generate effective descriptions of content within images and video clips.
Learn how to train a network using TensorFlow and the Microsoft Common Objects in Context (COCO) dataset to generate captions from images and video by:
Implementing deep learning workflows like image segmentation and text generation
Comparing and contrasting data types, workflows, and frameworks
Combining computer vision and natural language processing
Upon completion, you’ll be able to solve deep learning problems that require multiple types of data inputs.

Wednesday 6 February

09:00 → 10:45 Data Analytics

In this lesson we will introduce the ideas of Machine Learning. First we will introduce a set of general concepts, then we will try to describe and understand the basic ideas of classification and regression, and describe the principal techniques and ideas behind each technique.
In the afternoon we will then use these ideas in practice analysing a couple of different datasets for both classification and regression problems.

10:45 → 11:05 Coffe Break

11:05 → 12:50 Data Analytics

In this lesson we will introduce the ideas of Machine Learning. First we will introduce a set of general concepts, then we will try to describe and understand the basic ideas of classification and regression, and describe the principal techniques and ideas behind each technique.
In the afternoon we will then use these ideas in practice analysing a couple of different datasets for both classification and regression problems.

12:50 → 14:00 Lunch

14:00 → 15:45 Data Analytics

In this lesson we will introduce the ideas of Machine Learning. First we will introduce a set of general concepts, then we will try to describe and understand the basic ideas of classification and regression, and describe the principal techniques and ideas behind each technique.
In the afternoon we will then use these ideas in practice analysing a couple of different datasets for both classification and regression problems.

15:45 → 16:05 Coffe Break

16:05 → 17:50 Data Analytics

In this lesson we will introduce the ideas of Machine Learning. First we will introduce a set of general concepts, then we will try to describe and understand the basic ideas of classification and regression, and describe the principal techniques and ideas behind each technique.
In the afternoon we will then use these ideas in practice analysing a couple of different datasets for both classification and regression problems.

Thursday 7 February

09:00 → 10:45 Data Analytics

In this lesson we will describe time series: data that are “anchored” to a specific time. We will briefly introduce the principal ways of analysis modelling and forecasting of time series. We will also introduce a few theoretical ideas on time series and their analysis.
In the afternoon we will put these ideas into practice analysing a few different time series.

10:45 → 11:05 Coffe Break

11:05 → 12:50 Data Analytics

In this lesson we will describe time series: data that are “anchored” to a specific time. We will briefly introduce the principal ways of analysis modelling and forecasting of time series. We will also introduce a few theoretical ideas on time series and their analysis.
In the afternoon we will put these ideas into practice analysing a few different time series.

12:50 → 14:00 Lunch

14:00 → 15:45 Data Analytics

In this lesson we will describe time series: data that are “anchored” to a specific time. We will briefly introduce the principal ways of analysis modelling and forecasting of time series. We will also introduce a few theoretical ideas on time series and their analysis.
In the afternoon we will put these ideas into practice analysing a few different time series.

15:45 → 16:05 Coffe Break

16:05 → 17:50 Data Analytics

In this lesson we will describe time series: data that are “anchored” to a specific time. We will briefly introduce the principal ways of analysis modelling and forecasting of time series. We will also introduce a few theoretical ideas on time series and their analysis.
In the afternoon we will put these ideas into practice analysing a few different time series.

Friday 8 February

09:00 → 10:30 students presentations

10:30 → 11:00 Coffe Break

11:00 → 12:30 students presentations

12:30 → 14:00 Lunch

14:00 → 17:00 students presentations

Monday 11 February

09:00 → 10:45 Data storage and management

During the sessions of "Data Storage and Management", a foundation on data management for parallel scientific application is presented.

We will start with a generic introduction to formatted I/O and then focusing on parallel I/O. The final part of the workshop is aimed to show best practices of data production and data management for real applications, including an introduction to data visualization.

The course is mostly based on hands-on sessions, for learning the experience of a practical approach.

10:45 → 11:05 Coffe Break

11:05 → 12:50 Data storage and management

During the sessions of "Data Storage and Management", a foundation on data management for parallel scientific application is presented.

We will start with a generic introduction to formatted I/O and then focusing on parallel I/O. The final part of the workshop is aimed to show best practices of data production and data management for real applications, including an introduction to data visualization.

The course is mostly based on hands-on sessions, for learning the experience of a practical approach.

12:50 → 14:00 Lunch

14:00 → 15:45 Data storage and management

During the sessions of "Data Storage and Management", a foundation on data management for parallel scientific application is presented.

We will start with a generic introduction to formatted I/O and then focusing on parallel I/O. The final part of the workshop is aimed to show best practices of data production and data management for real applications, including an introduction to data visualization.

The course is mostly based on hands-on sessions, for learning the experience of a practical approach.

15:45 → 16:05 Coffe Break

16:05 → 17:50 Data storage and management

During the sessions of "Data Storage and Management", a foundation on data management for parallel scientific application is presented.

We will start with a generic introduction to formatted I/O and then focusing on parallel I/O. The final part of the workshop is aimed to show best practices of data production and data management for real applications, including an introduction to data visualization.

The course is mostly based on hands-on sessions, for learning the experience of a practical approach.

Tuesday 12 February

09:00 → 10:45 Data storage and management

During the sessions of "Data Storage and Management", a foundation on data management for parallel scientific application is presented.

We will start with a generic introduction to formatted I/O and then focusing on parallel I/O. The final part of the workshop is aimed to show best practices of data production and data management for real applications, including an introduction to data visualization.

The course is mostly based on hands-on sessions, for learning the experience of a practical approach.

10:45 → 11:05 Coffe Break

11:05 → 12:50 Data storage and management

During the sessions of "Data Storage and Management", a foundation on data management for parallel scientific application is presented.

We will start with a generic introduction to formatted I/O and then focusing on parallel I/O. The final part of the workshop is aimed to show best practices of data production and data management for real applications, including an introduction to data visualization.

The course is mostly based on hands-on sessions, for learning the experience of a practical approach.

12:50 → 14:00 Lunch

14:00 → 15:45 Data storage and management

During the sessions of "Data Storage and Management", a foundation on data management for parallel scientific application is presented.

We will start with a generic introduction to formatted I/O and then focusing on parallel I/O. The final part of the workshop is aimed to show best practices of data production and data management for real applications, including an introduction to data visualization.

The course is mostly based on hands-on sessions, for learning the experience of a practical approach.

15:45 → 16:05 Coffe Break

16:05 → 17:50 Data storage and management

During the sessions of "Data Storage and Management", a foundation on data management for parallel scientific application is presented.

We will start with a generic introduction to formatted I/O and then focusing on parallel I/O. The final part of the workshop is aimed to show best practices of data production and data management for real applications, including an introduction to data visualization.

The course is mostly based on hands-on sessions, for learning the experience of a practical approach.

Wednesday 13 February

09:00 → 10:45 Data storage and management

During the sessions of "Data Storage and Management", a foundation on data management for parallel scientific application is presented.

We will start with a generic introduction to formatted I/O and then focusing on parallel I/O. The final part of the workshop is aimed to show best practices of data production and data management for real applications, including an introduction to data visualization.

The course is mostly based on hands-on sessions, for learning the experience of a practical approach.

10:45 → 11:05 Coffe Break

11:05 → 12:50 Data storage and management

During the sessions of "Data Storage and Management", a foundation on data management for parallel scientific application is presented.

We will start with a generic introduction to formatted I/O and then focusing on parallel I/O. The final part of the workshop is aimed to show best practices of data production and data management for real applications, including an introduction to data visualization.

The course is mostly based on hands-on sessions, for learning the experience of a practical approach.

12:50 → 14:00 Lunch

14:00 → 15:45 Data storage and management

During the sessions of "Data Storage and Management", a foundation on data management for parallel scientific application is presented.

We will start with a generic introduction to formatted I/O and then focusing on parallel I/O. The final part of the workshop is aimed to show best practices of data production and data management for real applications, including an introduction to data visualization.

The course is mostly based on hands-on sessions, for learning the experience of a practical approach.

15:45 → 16:05 Coffe Break

16:05 → 17:50 Data storage and management

During the sessions of "Data Storage and Management", a foundation on data management for parallel scientific application is presented.

We will start with a generic introduction to formatted I/O and then focusing on parallel I/O. The final part of the workshop is aimed to show best practices of data production and data management for real applications, including an introduction to data visualization.

The course is mostly based on hands-on sessions, for learning the experience of a practical approach.

Thursday 14 February

09:00 → 10:30 Quantum technologies for future scientific research

In this lesson, we briefly review the basics of quantum information theory and some of the most promising applications of quantum technologies for future research. In particular, we present the achievements and the challenges obtained by quantum simulators, dedicated quantum hardware built to simulate interesting but hardly accessible physics: from models to study for high-Tc superconductors or topological systems, critical systems, quantum chemistry or lattice gauge theories where Monte Carlo methods efficiency is hindered by the sign problem. In particular, we present the first experimental quantum simulations of lattice gauge theories. Alongside, we present the tensor network methods, a powerful classical numerical approach that promise to become a powerful tool accompanying future quantum simulations and computations, providing guidance, benchmarking and verification of the quantum results.

Speaker: Simone Montangero (Universita' di Padova)

10:30 → 11:00 Coffe Break

11:00 → 12:30 Scalable analysis of ultra-terabyte brain images: from low-level data management to deep learning

Deciphering brain architecture at a system level requires the ability to quantitatively map its structure with cellular and subcellular resolution. Besides posing significant challenges to current optical microscopy methods, this ambitious goal require the development of a new generation of tools to make sense of the huge amount of raw images generated, which can easily exceeds several TeraBytes for a single sample. We present an integrated pipeline allowing the image transformation from a collection of voxel gray levels to a semantic representation of the sample.
As a first step, the hundreds of adjacent tiles produced by the microscope needs to be aligned and fuse together. To this aim, we developed ZetaStitcher, a software for image stitching that computes global optimal alignment of imaging datasets as big as 8 TB in less than an hour. The fused volume is then generated virtually, without the need to create a physical copy of the dataset, by means of a dedicated API.
The virtually fused volume is then processed to extract meaningful information. We demonstrate two complementary approaches based on deep convolutional networks. In one case, a 3D conv-net is used to ‘semantically deconvolve’ the image [Frasconi et al., Bioinformatics 2014], allowing accurate localization of neuronal bodies with standard clustering algorithms (e.g. mean shift). The scalability of this approach is demonstrated by mapping whole-brain spatial distribution of different neuronal populations with single-cell resolution.
To go beyond simple localization, we exploited a 2D conv-net estimating for each pixel the probability of being part of a neuron [Mazzamuto et al., LNCS 2018]. The output of the net is then processed with a contour finding, obtaining reliable segmentation of cell morphology. This information can be used to classify neurons, expanding the potential of chemical labeling strategies.

Speaker: Ludovico Silvestri (Istituto Nazionale di Ottica (INO-CNR))

12:30 → 14:00 Lunch

14:00 → 15:30 ECMWF's Extreme Data Challenges for Exascale Numerical Weather Prediction

Starting 2014, ECMWF has embarked on a 10 year research programme on HPC Scalability, aiming to achieve Exascale numerical weather prediction systems by 2025. ECMWF operational forecast generates massive amounts of I/O in short bursts, accumulating to tens of TiB in hourly windows. From this output, millions of user-defined daily products are generated and disseminated to member states and commercial clients all over the world.

These products are processed from the raw output of the IFS model, within the time critical path and under strict delivery schedule. Upcoming rise in resolution and growing popularity will increase both the size and number of these products. Based on expected model resolution upgrades, by 2020 we estimate the operational model will output over 100 TiB/day and need to archive over 400 TiB/day. Given that the I/O workload is already one of the strongest bottlenecks in ECMWF's workflow, this is one of the main challenges to reach Exascale NWP.

We present the latest ECMWF developments in model I/O and product generation, and how we are reworking our operational workflows to adapt to forthcoming new architectures and memory-storage hierarchies.

Speaker: Tiago Quintino (ECMWF)

15:30 → 16:00 Coffe Break

16:00 → 17:30 The challenge of Big Data in Science

In the last decade science has tackled problems in advanced research that required and provided a rapidly growing amount of data.
Along with the size of these data, also their complexity has grown exponentially, as well as the time required to process and transport them across research centers worldwide. Availability of these data, their storage, the necessary computing power to extract meaning, the network to share them are all problems intertwined inextricably. Moreover, new technologies have surfaced, like multithreading, vectorization (and parallelism in general), along with methodologies such as Neural Networks, Artificial Intelligence and Quantum Computing. At the same time new computing languages are emerging, to cope with the necessity of allowing people to express complex constructs and algorithms in a natural and efficient way.
In this seminar we will highlight a global picture of this emerging scenario and discuss the point of view of a researcher in this context.

Speaker: Dario Menasce (INFN Milano Bicocca)

Friday 15 February

09:00 → 10:30 students presentations

10:30 → 11:00 Coffe Break

11:00 → 12:30 students presentations

12:30 → 14:00 Lunch

14:00 → 15:30 students presentations

15:30 → 16:00 Coffe Break

16:00 → 17:30 students presentations