Luc Caspar, Msc/Ing.


  • Programming
    Java 3 years
    Java EE 3 years
    C/C++ 2 years
    Python 2/3 5 years
    Bash 5 years
  • Frameworks
    Numpy / SciPy 2 years
    DEAP 1 years
    Thespian 3 years
    Tensorflow 1 years
    Keras 1 years
    OpenAi Gym 1 years
    HDF5 / h5py 1 years
  • Web Development
    Html 4/5 6 years
    Php 7 5 years
    Css 2/3 6 years
    Javascript 6 years
    Ajax 6 years
    JQuery 6 years
    REST Api 7 years
    Backbone.js 5 years
    Django / Django REST 2 years
    Websockets 2 years
  • Integrated Development Environment
    PyCharm 4 years
    Vim / NeoVim 10 years
  • Office
    Latex 6 years
    LibreOffice 10 years
  • Robotics
    Robot Operating System (ROS) 2 years
    Webots 3 years
    E-Puck robot 3 years
    Nao robot 2 years
    Naoqi 2 years
    Sphero 1 years
    Gazebo 2 years
  • Design
    Inkscape 3 years
    Gimp 2.x 7 years
    Dia 4 years
  • System Administration
    Apache 2.x 5 years
    Arch Linux 10 years
    Ubuntu Desktop / Server 10 years
  • Machine Learning
    Reinforcement Learning 4 years
    Actor - Critic 3 years
    TD-/Q-Learning 4 years
    Genetic Algorithm 4 years
    Artificial Neural Network 6 years
    LEABRA / Emergent 3 years
  • Databases
    MySql 8 years
    MariaDB 8 years
  • Advanced Message Queuing Protocol
    RabbitMq 1 years
    Celery 1 years
  • Containers
    Docker 1 years
    Singularity 1 years
  • Miscellaneous
    Mendeley 4 years


Developer, The University of Sheffield

2015-10 — 2016-01 | Python Nao robot Naoqi Git

The position consisted in developing material and writing documentation for the laboratory sessions of the COM1005 module.

Software Engineer, Omnilog SA

2014-04 — 2014-08 | Php Javascript Ajax JQuery Html SQL Css Linux NodeJs

As a member of the maintenance team for the newspaper entitled: "Le Monde", my main mission was to fix any problems that might arise within the system. For each problem solved, documentation had to be written, detailing the steps taken to fix the issue. Hence, building a knowledge base, that would help future employee in a similar situation.

Software Engineer, Artaud, Courthéoux & associés (ACA)

2013-07 — 2014-02 | SQL Java Java EE Javascript Ajax JQuery REST Api Html Css Java Beans

Part of a small team of five individuals, our main mission was to develop a Web application for creating and managing insurance contracts and their associated customer accounts.

Co-founder, Creatix-web

2010-08 — 2014-02 | SQL Php Html Css Javascript JQuery Ajax Backbone.js REST Api Gimp 2.x Apache 2

A small business created, with Nicolas Ilhe (a classmate), during the summer between the first and second years of engineering school. Creatix Web was focused on creating complete websites, web applications and web designs. It also provided web hosting services, for small businesses. The main objective of this endeavor was to gain experience in Web programming, as well as management and client relationship.


PhD degree, The University of Sheffield

2014-09 — Present

Currently working on a project entitled: "The Role of Emotions in Autonomous Social Agents". This research project sits at the crossroad of neuroscience, robotics, artificial intelligence and affective science.

Language institute, chronologize

2012-09 — 2013-09

Basic and intermediate level Japanese training courses.

Master degree, Osaka Prefecture University (OPU)

2011-09 — 2012-09

A partnership between the EISTI and Osaka Prefecture University, allowed me to spend a year abroad and earn a double degree: A master in computer science and intelligent system, as well as the engineering degree.

Engineer diploma, Graduate school in computer science and mathematics engineering (EISTI)

2009-09 — 2012-09

Private engineering school, specialized in data processing and mathematics.

Preparatory Classes for Engineering Schools (CPGE), 806-384-5776

2007-09 — 2009-09

Part of the post-secondary education system. These two intensive years provide the knowledge and training required to take the enrollment exams for the different engineering schools.

Scientific baccalaureat option engineering sciences, Lycée Louis Marchal

2004-09 — 2007-06

Followed the standard high-school curriculum, with an emphasis on mechanic, electronic and programming.


  • French
  • English
    17 years, TOEFL: 105, TOEIC: 900
  • Japanese
    2 years
  • German
    7 years


The Role of Emotions in Autonomous Social Agents, The University of Sheffield


Affective science describes the field interested in the concept of emotions. Even though, it is concerned with the study of emotions, researchers in the field have yet to agree upon a standard definition for this concept. This fact, however, has not stopped the philosophers of old and more recently researchers to formulate myriads of theories and models. Furthermore, after being pitched against each other for centuries, cognitive science now informs how emotions are investigated. Resulting in cognitive architectures being usually either devoid of any emotions or treat them as yet another cognitive module, a sort of “magical” black box. Here we show that by focusing on emotions and the functions they evolved to fulfill, it is possible to build control systems endowed with survival capabilities and implementing emotional strategies. Consequently, the sub-cortical structures, thought to be hard wired meshes, are interpreted as mechanisms detecting relevant stimuli. The ProtoEmo architecture computing the salience of incoming stimuli and capable of adaptation via associative learning, therefore supports a primitive form of emotions. Its further integration with the dopamine and serotonin systems, resulting in the PrimEmo architecture, extends its capabilities to that of a critic able to inform the decision-making process. This allows PrimEmo to support basic emotions characterized by their valence and salience, all the while endowing agents and animals with survival abilities. Owing to its ability to learn from relevant stimuli and associated outcomes, PrimEmo is a robust basis upon which more complex cognitive architectures can be built. Future computational neuroscience projects could also aim at extending PrimEmo’s capabilities. Hence, providing an interpretation for brain systems supporting higher level cognitions interacting with or influenced by full-fledged emotions. It has recently been accepted that emotions are an essential component to general intelligence.

PrimEmo: A Neural Implementation of Survival Circuits Supporting Primitive Emotions, Proceedings of AISB Annual Convention 2017


Affective and cognitive sciences are both fields interested in the inner workings of the brain. While affective science focuses on the concept of emotions and how they are produced, cognitive science considers the brain as a whole, treating it as a system made of a multitude of independent subsystems. Since its inception, affective science has produced a plethora of theories and models each trying to solve the mysteries surrounding the definition and origin of emotions in the brain. Cognitive science on the other hand has had a rather chaotic history shifting its focus every decade, before finally being influenced by computer science and adopting the point of view of the brain as the most elaborate computational device. In this point of view, the gray matter residing within our skull is reduced to a system that takes sensory information on its inputs, processes it and outputs more data or actions. Each field has, more or less, evolved independently from one another so far, but both are now facing fundamental problems. On the one hand, the concept of emotions has yet to be completely defined and modeled. On the other, cognitive science is still trying to produce architectures imbuing artificial agents with human-level intelligence. This paper introduces a neural architecture based on the “survival circuits” framework (LeDoux, 2012) supporting primitive emotions and providing survival skills to artificial agents. Side-stepping the problem of defining the concept of emotions, the suggested neural structure focuses on identifying and modeling parts of the brain involved in survival functions (defense, thermo-regulation, maintenance of energy, nutritional supplies, reproduction and fluid balance). The neural implementation of this system provides a proto-brain for groups of artificial agents trying to survive in a dynamic virtual environment. By comparison with a hard-coded control logic (Scheutz, 2004), our architecture allows for a more equitable sharing of the resources and a longer life expectancy. It is our belief that, in the long term, the system suggested in this paper could become a robust basis upon which more elaborate cognitive architectures, such as ACT-R or SOAR, could be built, hence moving one step closer to endowing artificial agents with human-level intelligence.

Introduction of Emotions in Coevolving Multi-agent System,


Even if robots are built more and more intelligent, they still lack one important ability: having Emotion. Our research focuses on how to represent emotions in computer science, an important topic in nowadays research. We apply this concept to learning robots that have to solve complex problems in cooperation. Because emotion is something dynamic, we also apply the concept of evolution to these robots. After some simulations on a co-evolution problem, we observed that the learning and evolving robots perform better than the other.

Emotion and Coevolution of Learning Robots Teams,


Emotion in robotic is an important subject in current researches. My work consists in building a framework for multi-agent system solving co-evolution problems using emotions. In a simulated environment learning robots, sharing knowledge, emotions and evolving, have to put their efficiency in solving problems in cooperation on display. Simulation results show that by sharing knowledge and emotions a team of robots is able to achieve its goal faster and thanks to the emotions’ representation the amount of data exchanged is reduced to its minimum.


Talk, AISB Conference


A presentation detailing the ProtoEmo architecture, which was developed during my PhD project. The talk also described the simulation results gathered using ProtoEmo as a control mechanism for groups of virtual agents, having to survive in a dynamic environment.

Poster, EuCog Conference


Presented a poster introducing the basic concepts upon which the ProtoEmo architecture had been developed. The poster also contained some preliminary simulation results showing the potential of the ProtoEmo architecture as a control/decision-making system.

Talk, SICE Conference


An extended presentation of the results gathered during my master project at Osaka Prefecture University.

Talk, SCI Conference


A presentation describing the model of emotions developed for my master project. The talk also included some preliminary results showing the potential of using emotions in teams of co-evolving robots.