Page 10 - Study At Centrale Supélec 2021

Computer Science

ENGINEERING CHALLENGE TERM ACADEMIC TERM

CORE COURSE : OPTIMISATION (2,5 ECTS)
(MANDATORY) ELECTIVE SERIES 2.4 (2,5 ECTS)
This course will explore various fundamental notions of both continuous and • ARTIFICIAL INTELLIGENCE
discrete optimisation. The following topics will be addressed and implemented : What do web-based information retrieval, personal assistant development,
formulation of optimisation problems, existence conditions for global and local autonomous driving or automatic planning have in common? These are all
minimizers, convexity, duality, Lagrange multipliers, first-order methods, linear complex real-world problems that artificial intelligence (AI) aims at solving
programming, integer linear programming, branch and bound approaches, by addressing them in rigorous methods. In this course, you will study the
preliminary stochastic optimisation concepts. fundamental principles that guide these applications and implement some of
these systems. Specific topics include automated learning, research, gaming,
CONCENTRATION (7,5 ECTS) Markov decision processes, constraint satisfaction, graphic models and logic.
The main objective of the course is to provide you with the framework to
address new AI problems that you may encounter in the future. The ethical and
• HIGH PERFORMANCE SIMULATION FOR FOOTPRINT philosophical aspects of AI will also be discussed.
REDUCTION
ELECTIVE SERIES 2.5 (2,5 ECTS)
Concentration course : Parallel Computing Methods
and Algorithms, and Optimisation Methods • INTERACTIVE ROBOTIC SYSTEMS
PARIS SACLAY to reduce the impact of the products created: reducing the carbon footprint, complex systems in interaction with humans or their environment. The course
Subjects covered by this course will allow students to understand the main
issues of interactive robotics and the technical aspects associated to these
Simulation today is at the heart of many design and optimisation approaches,
the sound footprint, etc. Such problems are often complex systems, whose
aims at exposing the context, fundamental methodological tools and current
research and development subjects related to interactive robotic manipulators.
simulation requires specific skills in high performance and large-scale
simulations.
In this course, students will learn to develop models and simulations whichever
ELECTIVE SERIES 2.6 (2,5 ECTS) (SELECT ONE)
the size of the problem, without sacrificing the accuracy of computations.
Concentration projects (SELECT ONE) • OBJECT ORIENTED SOFTWARE ENGINEERING
Software Engineering (SE) is a discipline concerned with concepts, techniques
and tools aimed at the production of quality software. This course aims at
Project (1) : Energy optimisation and acceleration of a providing engineering students with an overview of the problem of software
cloud financial calculation graph design and development by means of the object-oriented programming (OOP)
In partnership with ANEO paradigm. By learning the Java programming language, students will acquire
Modern insurers have a highly regulated but at the same time relatively broad basic skills in the software development process using a state-of-the-art
SEMESTER BY COURSEWORK
field of activity : different types of insurance, banking services, etc. One of the Integrated Development Environment (IDE). By focusing on object-oriented
difficulties in assessing the accounts of an insurance company (or bank) lies in modelling, the UML language, the Javadoc-based code documentation,
the valuation of financial assets and the underlying risks. In order to optimise the Junit-based development of unit-tests, students will acquire basic skills
costs without investing in a very large computing grid that would ultimately be essential to the realisation of industrial software.
little bing the infrastructure scheduling (switching on and off nodes) as well as
the task scheduling (placement of a task on a node at a given time). • APPLICATIONS OF STATISTICAL PHYSICS TO
INFORMATION PROCESSING
Project (2) : Low cost optimisation of acoustic wave This transdisciplinary course aims at establishing connections between basic
propagation code performance mathematics and physics training and applications in the advanced technology
In partnership with INTEL areas, such as digital communications, data processing, learning, and quantum
Regardless of the type of application running on parallel machines in an HPC computing. This course will also be a necessary opening for students wishing
(high performance computing) environment, and regardless of their level of to be prepared for modern computational techniques and for research
efficiency (in terms of performance or energy footprint), we can easily see and engineering in relevant fields such as: theoretical physics, quantum
that the impact of input parameters is generally not negligible. It is therefore computing, computational biology, neuroscience, telecommunications, artificial
necessary to choose optimization methods that do not consume too many intelligence, Big Data, social sciences, finance, etc.
experiments, and then to optimize their use to reduce the footprint of the
targeted HPC code, without this pre-study consuming too many computing
resources ! This amounts to « looking for the least expensive optimization
method » to find an optimum between convergence speed (of the optimization
algorithm) and the quality of the minimization of execution time and energy
footprint (of the targeted HPC application).

10