Prédiction, gestion et optimisation de la performance énergétique basées sur l'apprentissage automatique en vue d'améliorer le processus de prise de décision en matière énergétique
Context
-
Buildings are in charge of more than 40% of power consumption demand and greenhouse gas emissions
-
Need for a better understanding, efficient characterization and prediction of the energy building performance at different scales for informed decision-making to reduce energy costs and consumption
-
Multi-source data (sensors, smart meters, numerical simulations, weather data…)
Méthodes de modélisation hybrides vers une ingénierie augmentée - Use Case : Fabrication additive assistée par IA
Context
- Goal of Numerical Simulation and Machine Learning : Predict the behaviour/performance of a system/product/process
- Data can be rather expensive
- Specific challenge of small but smart datasets
→ Frugal learning - Combine simulation and ML towards hybrid modeling approaches & Digital Twins
- Vast amount of domain prior knowledge is available
TRAIL Researcher's Charter
You will find below the "TRAIL researcher's charter" with a version in French and another in English.
Charte du chercheur TRAIL
Vous trouverez ci-dessous la "Charte du chercheur TRAIL" avec une version en français et une autre en anglais.
An exploration of the performances achievable by combining unsupervised background subtraction algorithms
Background subtraction (BGS) is a common choice for performing motion detection in video. Hundreds of BGS algorithms are released every year, but combining them to detect motion remains largely unexplored. We found that combination strategies allow to capitalize on this massive amount of available BGS algorithms, and offer significant space for performance improvement. In this paper, we explore sets of performances achievable by 6 strategies combining, pixelwise, the outputs of 26 unsupervised BGS algorithms, on the CDnet 2014 dataset, both in the ROC space and in terms of the F1 score.
Leveraging Predictions from Multiple Repositories to Improve Bot Detection
Contemporary social coding platforms such as GitHub facilitate collaborative distributed software development. Developers engaged in these platforms often use machine accounts (bots) for automating effort-intensive or repetitive activities. Determining whether a contributor corresponds to a bot or a human account is important in socio-technical studies, for example, to assess the positive and negative impact of using bots, analyse the evolution of bots and their usage, identify top human contributors, and so on.
Federated Inductive Logic Programming
It is a commonly accepted fact that machine learning requires large amounts of data. Fortunately, the sources of information are more and more numerous and the amount of data available in all domains is constantly increasing. However, this evolution has reached a point where it is no longer realistic to think of storing the whole set of data needed for a machine learning task on a single computer. This has led J. Konecny, H.B. MacMahan and D. Ramage to propose a new learning model in which the data is scattered on distributed nodes and the model is learned in a distributed manner.
User-centric XAI and Visualization tools
Objectives:
Enhance the trustworthiness of nonlinear dimensionality reduction algorithms by developing explainability methods for low-dimensional (typically 2 or 3-D) nonlinear embeddings, as analyzed in the context of exploratory data visualization tasks.
Research RoadMap:
- Review of the state of the art: current explainability techniques for nonlinear embeddings.
- Identification of their limitations.
- Design of an explainability framework to address them.