CPS2: AI and IoT is the international track of the Master in Computer
Science at Université de Lyon, jointly operated by Université Jean
Monnet Saint-Étienne and Mines Saint-Étienne (Institut Mines-Télécom).
Research Interests
My research interests represented as a concept map
Research & Projects
SemanticGAN: Knowledge Graph Completion with Wasserstein GANs
An experimental research system for knowledge graph completion and
RDF triple generation.
A Systematic Mapping Study on Green Load Balancing Algorithms for
Cloud Data Centers
A systematic mapping study that reviews published research on load
balancing in cloud data centers. The paper classifies existing work
and discusses open issues related to energy use and resource
allocation.
Safety Critical Systems in Medical Cyber-Physical Systems
A report that studies safety challenges in medical cyber physical
systems through a real medical device recall case. The work reviews
existing safety approaches and discusses how design time
verification, runtime mechanisms, and system architecture can be
combined to improve patient safety.
A report that examines data interoperability issues in Sweden’s
national open data portal using parliamentary and public datasets.
The work analyzes the use of metadata standards and shows how
semantic models can be used to align heterogeneous data schemas.
A compendium (pl. compendia or compendiums) is a comprehensive collection of information and analysis pertaining to a body of knowledge. A compendium may concisely summarize a larger work. In most cases, the body of knowledge will concern a specific field of human interest or endeavor (for example: hydrogeology, logology, ichthyology, phytosociology or myrmecology), while a general encyclopedia can be referred to as a "compendium of all human knowledge".
Markov Decision Processes (MDP)
The formalism for Reinforcement Learning problems consisting of:
States (s), Actions (A), Transition Model T(s,a,s') which is the
probability Pr(s'|s,a), and Rewards R(s).
Bias and Variance Tradeoff
Bias is how well the model hits its target. Variance is how
consistent the model is across multiple datasets. Methods to
decrease Bias generally increase Variance (no free lunch theorem).
Ensemble Learning
One method to decrease Bias. Various types: Bagging Bootstrap
(select data) then Aggregate (avg) identical models. Stacking Learn
different models to Aggregate with vote/another model. Boosting Use
previous errors to improve the (same) model.
Hebbian Learning
When an axon of cell A is near enough to excite a cell B and
repeatedly or persistently takes part in firing it, some growth
process or metabolic change takes place in one or both cells such
that A's efficiency, as one of the cells firing B, is increased.
A System on a Chip (SoC) is an integrated circuit (IC) that
integrates all or most components of a computer or other electronic
system. These components almost always include a central processing
unit (CPU), memory, input/output ports and secondary storage.
LPWAN
Low Power Wide Area Network (LPWAN) is a type of wireless
telecommunication wide area network designed to allow long range
communications at a low bit rate among things (connected objects).
LoRaWAN
LoRaWAN is a media access control (MAC) layer protocol for managing
communication between LPWAN gateways and end-node devices,
maintained by the LoRa Alliance.
Matter Standard
Matter is a proprietary, royalty-free home automation connectivity
standard. It aims to reduce fragmentation across different vendors
and achieve interoperability among smart home devices and Internet
of things (IoT) platforms from different providers.
Software Requirement
Software capability needed by a user to solve a problem or to
achieve an objective. Software capability that must be met or
possessed by a system or system component to satisfy a contract,
standard, specification, or other formally imposed document.
Unified Modeling Language (UML)
The UML is a general-purpose visual modeling language intended to
provide a standard way to visualize, specify, construct and document
the artifacts of a software-intensive system.
Software Product Quality
Degree to which the system satisfies the stated and implied needs of
its various stakeholders, and thus provides value.
DevOps
DevOps is a set of practices that combines software development
(Dev) and IT operations (Ops). It aims to shorten the systems
development life cycle and provide continuous delivery with high
software quality.
Scrum Framework
A Scrum Master fosters an environment where: 1. A Product Owner
orders the work for a complex problem into a Product Backlog. 2. The
Scrum Team turns a selection of the work into an Increment of value
during a Sprint. 3. The Scrum Team and its stakeholders inspect the
results and adjust for the next Sprint. 4. Repeat.
Semantic Data Interoperability
Interoperability so that the meaning of the data model within the
context of a subject area is understood by the participating
systems.
FAIR Principles
The FAIR principles emphasize machine-actionability (i.e., the
capacity of computational systems to find, access, interoperate, and
reuse data with none or minimal human intervention) because humans
increasingly rely on computational support to deal with data as a
result of the increase in volume, complexity, and creation speed of
data.
Common European Data Spaces
Purpose or sector-specific or cross-sectoral interoperable
frameworks for common standards and practices to share or jointly
process data for, inter alia, the development of new products and
services, scientific research or civil society initiatives.
Controlled Vocabularies
Vocabulary for which the entries, i.e. definition/term pairs, are
controlled by a Source Authority based on a rulebase and process for
addition/deletion of entries.
