Emplois actuels liés à Stage In-Painting Using Generative ai For The Evaluation Of Explainable ai Xai Methods H/F - Grenoble - CEA
-
Stage: In-painting Using Generative Ai for The
Il y a 2 mois
Grenoble, France CEA Temps pleinPosition description **Category**: - Mathematics, information, scientific, software **Contract**: - Internship **Job title**: - STAGE: In-painting using generative AI for the evaluation of eXplainable AI (XAI) methods H/F **Subject**: - In-painting using generative AI for the correctness evaluation of eXplainable AI (XAI) methods H/F **Contract...
-
AI Researcher
il y a 3 semaines
Grenoble, Auvergne-Rhône-Alpes, France CEA Temps pleinJob Title: In-painting using Generative AI for XAI EvaluationWe are seeking a highly skilled researcher to join our team at CEA LIST Software Safety and Security Laboratory (LSL) to work on a project focused on evaluating the correctness of Explainable AI (XAI) methods using generative AI.Job Description:The successful candidate will be responsible for...
-
Stage: Self-explainable Model for Audio
Il y a 2 mois
Grenoble, France CEA Temps pleinPosition description **Category**: - Mathematics, information, scientific, software **Contract**: - Internship **Job title**: - STAGE: Self-explainable model for audio identification of bird species H/F **Subject**: - Self-explainable model for audio identification of bird species H/F **Contract duration (months)**: - 5-6 months **Job...
-
Stage - Out-of-distribution Detection for
Il y a 2 mois
Grenoble, France CEA Temps pleinDescription du poste **Domaine**: - Mathématiques, information scientifique, logiciel **Contrat**: - Stage **Intitulé de l'offre**: - STAGE - Out-of-distribution detection for adversarial attacks evasion H/F **Sujet de stage**: - During this internship, you will study the use of AISER's OOD-detection method, PARTICUL, to identify whether new inputs...
-
Research Intern
il y a 3 semaines
Grenoble, Auvergne-Rhône-Alpes, France CEA Temps pleinInternship OpportunityWe are seeking a highly motivated and talented research intern to join our team at CEA List, a leading research institute in the field of artificial intelligence and cybersecurity. The successful candidate will have the opportunity to work on a cutting-edge project focused on out-of-distribution detection for adversarial attacks...
-
Research Internship in AI Safety and Explainability
il y a 2 semaines
Grenoble, Auvergne-Rhône-Alpes, France CEA Temps pleinJob DescriptionThe CEA List institute is seeking a research intern to work on a project related to AI safety and explainability. The successful candidate will have the opportunity to work on a cutting-edge project that aims to develop methods for generating explanations of AI model behavior.The internship will involve studying the use of AISER's...
-
Modeling Expert
il y a 3 semaines
Grenoble, Auvergne-Rhône-Alpes, France CEA Temps pleinJob Title: Modeling ExpertWe are seeking a highly skilled Modeling Expert to join our team at CEA. As a Modeling Expert, you will be responsible for developing and implementing self-explainable models for audio classification.Job Description:Our team is working on a project to develop a self-explainable model for audio classification, specifically for bird...
-
Formal Methods for Distributed Systems Verification
il y a 3 semaines
Grenoble, Auvergne-Rhône-Alpes, France VERIMAG - Imag Temps pleinJob Title: Formal Methods for Distributed Systems VerificationVERIMAG - Imag is seeking a highly skilled candidate to join our team in the development of new verification techniques for distributed systems. The successful candidate will work on the verification of self-adapting distributed systems, focusing on formal methods and dynamic reconfiguration.Job...
-
Stage Bac +5 Modélisation Batteries Pinn
Il y a 2 mois
Grenoble, France CEA Temps pleinDescription du poste **Domaine**: - Electromagnétisme, génie électrique **Contrat**: - Stage **Intitulé de l'offre**: - Stage BAC +5 Modélisation batteries PINN H/F **Sujet de stage**: - Physics Informed Neural Networks (PINN) for Li-Ion battery modeling **Durée du contrat (en mois)**: - 6 **Description de l'offre**: - Missions: Physics...
-
Grenoble, France CEA Temps pleinDescription de l'offre Our team is dedicated to advancing the understanding of the complement system and its structural components, with a primary focus on autoimmune and chronic inflammatory diseases. Our current projects explore: - The interplay between alarmins and the classical complement pathway. - The novel roles of C1 proteases in the...
-
Development of a 1-6µm Optical Collection for
il y a 3 semaines
Grenoble, France CEA Temps pleinPosition description **Category**: - Engineering science **Contract**: - Internship **Job title**: - Development of a 1-6µm optical collection for cathodoluminescence H/F **Subject**: - The development of infrared imaging systems requires the creation of pixel arrays in structures based on small-gap materials (1-10µm) such as CdZnTe or HgCdTe...
-
Explanations of Optimal Policies for Markov
Il y a 2 mois
Grenoble, France CEA Temps pleinDescription du poste **Domaine**: - Mathématiques, information scientifique, logiciel **Contrat**: - Stage **Intitulé de l'offre**: - Explanations of Optimal Policies for Markov Decision Processes H/F **Sujet de stage**: - Explanations of Optimal Policies for Markov Decision Processes **Durée du contrat (en mois)**: - 5 to 6 **Description de...
-
Engineer in the Radio Frequency group
Il y a 5 mois
Grenoble, France ESRF Temps pleinContext & Job description The Accelerator and Source Division (ASD) operates an accelerator complex with a 6 GeV electron storage ring, which is the source of the X-rays that are exploited on the experimental stations. In 2020, the ESRF has commissioned a new ultra-low emittance ring that constitutes the world’s first Extremely Brilliant Source...
-
Grenoble, Auvergne-Rhône-Alpes, France ESRF Temps pleinInvestigate the Connection Between Deformation and Annealing MicrostructuresThis PhD position at the ESRF is part of the ERC Starting Grant Project "D-REX". The project aims to utilize advanced X-ray microscopy techniques at the ID03 beamline to study dislocation structures, subgrain evolution, and recrystallization in highly strained metals. As a PhD...
-
PhD Student on ID03 in the Structure of Materials group
Il y a 4 mois
Grenoble, France ESRF Temps pleinContext & Job description Thesis subject: Advanced 3D Characterization of Deformed and Recrystallized Microstructures in Highly Strained Metals As a PhD student, you will play a crucial role in the investigation of the connection between deformation and annealing microstructures in metals, addressing the challenge of mapping deformed and...
-
PhD Student in Artificial Enzyme Design
il y a 1 mois
Grenoble, Auvergne-Rhône-Alpes, France ESRF Temps pleinResearch Scientist in Artificial Enzyme DesignWe are seeking a highly motivated and skilled Research Scientist to join our team at the European Synchrotron Radiation Facility (ESRF) in Grenoble, France. The successful candidate will be part of a dynamic and international research team working on the design and engineering of artificial enzymes for green...
-
Grenoble, Auvergne-Rhône-Alpes, France CEA Temps pleinJob OfferThe CEA is seeking a postdoctoral researcher to work on the experimental and numerical characterization of the thermal-mechanical behavior of granular media. The project focuses on the development of a thermal energy storage system using a granular bed, and the postdoctoral researcher will be responsible for conducting experiments and numerical...
-
Security Characterization of a Risc-v Processor
il y a 1 semaine
Grenoble, France CEA Temps pleinDescription du poste **Domaine**: - Mathématiques, information scientifique, logiciel **Contrat**: - Stage **Intitulé de l'offre**: - Security characterization of a RISC-V processor H/F H/F **Sujet de stage**: - The French Alternative Energies and Atomic Energy Commission (CEA) is a key player in research, devel-opment and innovation in four main...
-
Post-doctoral fellow
Il y a 5 mois
Grenoble, France ESRF Temps pleinContext & Job description You will participate in the operation of the beamline, ongoing in-house research programmes, provide support to external users as local contact, and develop your own research programme. Together with the beamline scientists, you will be required to participate in the development of data analysis methods and sample...
-
Head of The Instrumentation and Control Group
Il y a 5 mois
Grenoble, France ILL Temps plein**Head of the Instrumentation and Control Group** **Context** The Institut Laue-Langevin (ILL), situated in Grenoble, France, is Europe's leading research facility for fundamental research using neutrons. The ILL operates the brightest neutron source in the world and hosts over 2000 visits by scientists per year, who come to carry out world-class...
Stage In-Painting Using Generative ai For The Evaluation Of Explainable ai Xai Methods H/F
Il y a 2 mois
Through the recent developments of AI, the use of models produced by machine learning has become widespread, even in industrial settings. However, studies are flourishing showing the dangers that such models can bring, in terms of safety, privacy or even fairness. To mitigate these dangers and improve trust in AI, one possible avenue of research consists in designing methods for generating *explanations* of the model behaviour. Such methods, regrouped under the umbrella term "eXplainable AI" (XAI), empower the user by providing them with relevant information to make an informed choice to trust the model (or not).
In the field of XAI, multiple metrics have been proposed to evaluate the correctness of an explanation, i.e. how well the explanation reflects the actual AI model behaviour. In the particular context of computer vision, most evaluation metrics from the state of the art propose to de-activate pixels (e.g. replacing them with a black pixels) to measure their impact on the model decision. However, recent work has shown that such metrics might not BE informative, in the sense that they tend to evaluate the model behaviour on images that do not belong the training distribution and that can BE considered as out of distribution : indeed, an image with entire regions painted in black can hardly BE considered as a "normal" input that the model should expect during its lifecycle.
In this internship, we propose to use generative AI to de-activate pixels in a more subtle way - creating images that resemble the original one but with missing features while remaining in distribution - and to study the impact of such method on the evaluation of the correctness of XAI methods.
More precisely, the internship will BE split in several subtasks as follows :
Establish a baseline of existing metrics for evaluating the correctness of XAI methods, using the Quantus framework.
Identify a body of existing works on the use of generative AI for in-painting and select a method based on a set of motivated criteria
Implement the selected method and evaluate the advantages and drawbacks of the resulting evaluation metric, compared to the state of the art
