Introduction

About the school

Italian Computer Science PhD granting institutions under the auspices of GRIN, organizes an annual school offering three graduate-level courses aimed at first-year PhD students in Computer Science. In addition to introducing students to timely research topics, the school is meant to promote acquaintance and collaboration among young European researchers. The 2020 edition of the School is the 26th in the series.
The school will offer 3 courses each consisting of 13 hours of lectures:

  • Variational Laws of Learning
    Marco Gori, Siena Artificial Intelligence Lab (SAILab) & University of Siena
  • Space-Efficient Data Structures for Algorithm Design
    Roberto Grossi, Department of Computer Science & University of Pisa
  • Program Verification with F*
    Catalin Hritcu, INRIA & Paris

A final evaluation for each course is possible through a final exam or project as determined by the instructor. The daily schedule admits laboratory and/or working group activities to be organized in addition to the lectures.
The registration fee for the School is 550.00 Euro and includes all local expenses from the evening of 24 May to 29 May afternoon including all meals and on-site lodging in double-occupancy rooms. A reduced registration fee of 300.00 Euro is available for local students which will not use the on-site lodging facilities (it includes local expenses for the lectures, coffee breaks and lunches). It is possible to require one additional night for the students that want to leave on Saturday the 28th. In this case the additional cost is 50 Euros.
Attendance is limited to 50 students and will be allocated on a first-come-first-served basis.

Courses

Variational Laws of Learning

Marco Gori, Siena Artificial Intelligence Lab (SAILab) & University of Siena

Abstract: By and large, Backpropagation (BP) is regarded as one of the most important neural computation algorithms at the basis of the progress in machine learning, including the recent advances in deep learning. However, its computational structure has been the source of many debates on its arguable biological plausibility. In this course, it is shown that when framing supervised learning in the Lagrangian framework, while one can see a natural emergence of Backpropagation, biologically plausible local algorithms can also be devised that are based on the search for saddle points in the learning adjoint space composed of weights, neural outputs, and Lagrangian multipliers. This might open the doors to a truly novel class of learning algorithms where, because of the introduction of the notion of support neurons, the optimization scheme also plays a fundamental role in the construction of the architecture.

Space-Efficient Data Structures for Algorithm Design

Roberto Grossi, Department of Computer Science & University of Pisa

Abstract: The lectures will describe some basic algorithms and data structures that have been developed in the last 20 years and are widely recognized to be at the core of efficient algorithms with very low space occupancy. As today lots of data are available in the form of text, semi-structured data, or linked data, these data structures address how to store sequences, texts, parentheses, trees, and graphs in compressed format, so that accessing a portion of them does not require to decompress the entire format. Some applications in real-world scenario will illustrate their usefulness.

Program Verification with F*

Catalin Hritcu, INRIA & Paris

Abstract: F* (https://www.fstar-lang.org) is a general-purpose functional programming language with effects aimed at program verification. It puts together the automation of an SMT-backed deductive verification tool with the expressive power of a proof assistant based on dependent types. After verification, F* programs can be extracted to efficient OCaml, F#, or C code. This enables verifying the functional correctness and security of realistic applications, such as a cryptographic library and a HTTPS stack. This course will give a gentle introduction to program verification in F* using simple examples and a few, even simpler exercises along the way. We will start with specifying and verifying purely-functional programs and then move to programs with side-effects such as divergence and mutable state. Finally, if there is time and interest, we could also have a quick at how F* works under the hood by outlining its main innovations, such as Dijkstra monads. Closer to the event I'll also send you some instructions for the students who want to install F* locally on their machine. We will also provide a web interface though, so if the Internet works well that's not strictly required.

Programmes


Variational Laws of Learning (VLL)
Space-Efficient Data Structures for Algorithm Design (SED)
Program Verification with F* (PV)

2020
24/05
25/05
26/05
27/05
28/05
29/05
Sun
Mon
Tue
Wed
Thu
Fri
08.00-09.00
breakfast
09.00-10.00 PV VLL VLL SED SED
10.00-11.00 PV VLL VLL SED SED
11.00-11.30 coffee break
11.30-12.30 VLL PV VLL SED SED
12.30-13.30 VLL PV VLL SED SED
13.30-15.00 lunch
15.00-16.00 PV VLL PV PV departure
16.00-17.00 PV VLL PV PV -
17.00-17.30 coffee break
17.30-18.30 arrival VLL PV SED SED -
18.30-19.30 VLL PV SED SED -

ORGANIZATION

Scientific Organizing Committee


Paolo Boldi, University of Milano
Pierpaolo Degano, University of Pisa
Maurizio Gabbrielli, University of Bologna

Local Organization


Andrea Bandini, CeUB
Monica Michelacci, CeUB
Michela Schiavi, CeUB

Registration

The registration fee for the School is 550.00 Euro and includes all local expenses from the evening of 24 May to 29 May afternoon including all meals and on-site lodging in double-occupancy rooms. A reduced registration fee of 300.00 Euro is available for local students which will not use the on-site lodging facilities (it includes local expenses for the lectures, coffee breaks and lunches). It is possible to require one additional night for the students that want to leave on Saturday the 28th. In this case the additional cost is 50 Euro. Attendance is limited to 50 students and will be allocated on a first-come-first-served basis.

In order to register, all applicants must fill the form available at the following link: REGISTRATION FORM.

Venue

BISS events are held in the University Residential Center located in the small medieval hilltop town of Bertinoro. This town is in Emilia Romagna about 50km east of Bologna at an elevation of 230m above sea level.

Via Frangipane, 6 in Bertinoro, Italy

+39 0543 446500

segreteria@ceub.it

http://www.ceub.it

Past Editions

BISS 2019:

  • Internet of things: a data oriented approach
  • Multitask learning and learning-to-learn: a statistical learning perspective
  • Software security across abstraction layers

BISS 2018:

  • Provable security for low level execution platforms
  • Distributed models, MapReduce and large scale algorithms
  • Elements of Quantum Computation

BISS 2017:

  • Approximation Algorithms
  • Probabilistic Graphical Models in Intelligent Systems
  • Kleene algebra with tests and applications to network programming

BISS 2016:

  • Advanced Topics in Programming Languages
  • Model and Languages for Service-Oriented and Cloud Computing
  • Algorithms and Methods for Mining Large Graphs

BISS 2015:

  • Game Theory: Models, Numerical Methods, and Applications
  • Protection of sensitive information
  • Introduction to Modern Cryptography

BISS 2014:

  • Big Data Analysis of Patterns in Media Content
  • An Introduction to Probabilistic and Quantum Programming
  • Development of dynamically evolving and self-adaptive software

BISS 2013:

  • Foundations of Security: Cryptography, Protocols, Trust
  • Stochastic Process Algebras for Quantitative Analysis
  • Shape and Visual Apperance Acquisition for Photo-realistic Visualization

BISS 2012:

  • Algorithms for the web and for social networks
  • Software Verification and Interactive Theorem Proving
  • Regularization methods for high dimensional learning

BISS 2011:

  • Computational Aspects of Game Theory
  • Trust in Anonymity Networks (TAN)
  • Information Integration (II)
  • Model Checking: From Finite-state to Infinite-state Systems (MCFIS)