Dear colleagues, We have a two-year postdoc position to fill in the EOS VeriLearn (Verifying Learning Artificial Intelligence Systems) project at the Faculty of Computer Science of UNamur (Namur, Belgium).
VeriLearn is a large project involving 10 academics from three Belgian unversities. Within VeriLearn, the candidate will be integrated in an interdisciplinary team working at the intersection of AI, ML and SE.
The goal of this postdoctoral research is to define, design and evaluate “active testing” (a mix of active learning and software testing) techniques for ML models.
In particular, the successful candidate will analyse automatically the datasets to infer properties of interest in the form of constraints, rank them depending on the desired validation goal and integrate them in testing techniques (metamorphic testing, constraint-guided fuzzing, etc.).
This candidate will work with Prof. Benoit Frénay and Dr. Gilles Perrouin in the Faculty of Computer Science at the University of Namur in collaboration with the KULeuven DTAI group.
More infos on https://jobs.unamur.be/emploi.2021-09-09.2912866089/view
https://euraxess.ec.europa.eu/jobs/681147
Deadline: October, 15th, 2021 (11h59 pm AoE). Expected starting date: January 1st, 2022 (negotiable, depending on COVID or candidate constraints).
Please send your application or questions to [email protected] and [email protected].
Best regards, Benoît Frénay Gilles Perrouin
Hello, Please find below an advertisement for a Master internship (6 months) on:
Contribution of autoencoders to measure the resemblance between visual patterns (AI, ecology, evolution - CNRS Montpellier)
Person proposing the internship : Julien RENOULT, CNRS Researcher, Biologist of evolution, cognitive sciences, AI applied to biology
Professional address: CNRS-CEFE, 1919 route de Mende 34293 MONTPELLIER
Phone number: 04.67.61.32.10
Email address: [email protected]
Company / Home unit : CNRS - Center for Functional and Evolutionary Ecology CEFE-UMR5175.
Key words : artificial intelligence, evolutionary biology, autoencoders, sparsity.
Subject description The resemblance between visual patterns is an important component of communication between animals, for example in the case of mimicry or camouflage.
Recent work in psychology has also shown that communication signals that mimic certain visual properties of natural habitats tend to be considered attractive.
For example, by comparing the scale invariance of the body colored patterns of several species of fish living in American rivers with the scale invariance of their respective habitats, we recently showed a positive correlation for males. during breeding season, but not for females.
The objective of this internship is to study, beyond the particular case of scale invariance, the resemblance between the visual patterns of communication signals and those of natural habitats using autoencoders (AE ). By training AEs to reconstruct habitat images (one AE per habitat type), we assume that images of male fish (but not females) of a given species will be encoded most efficiently (eg , high neural code sparsity) and best reconstructed (lower reconstruction errors) by AEs trained to reconstruct images of the habitat in which this species lives.
The student will test this hypothesis by studying different architectures of autoencoders (eg, classical convolutional AEs, Variational Auto Encoders) and different cost functions (eg, with or without regularization of sparsity during learning). He / she will also be able to test the usefulness of the method in contexts other than sexual selection (eg, camouflage).
The student should have a good level of programming in Python, basic knowledge of AI and an interest in biology. She / he will be immersed in the E3CO team which has several students (Master, Doctorant.es) in AI applied to biology. He / she will have access to a computer equipped with GPU cards.
Place of the internship : E3CO team, Center for Functional and Evolutionary Ecology (CEFE), Montpellier.
Date and duration of the course : adjustable according to the requirements of the course.
Gratuity : university flat rate (around 590 € / month)
Master 2 internship (6 months)
Contribution of CNNs to model the link between beauty and neuronal fluence (AI, cognitive sciences - CNRS Montpellier)
Person proposing the internship : Julien RENOULT, CNRS Researcher, Biologist of evolution, cognitive sciences, AI applied to biology
Professional address: CNRS-CEFE, 1919 route de Mende 34293 MONTPELLIER
Phone number: 04.67.61.32.10
Email address: [email protected]
Company / Home unit : CNRS - Center for Functional and Evolutionary Ecology CEFE-UMR5175.
Key words : artificial intelligence, experimental psychology, beauty modeling, information theory, sparsity, entropy.
Subject description Modeling beauty represents a timeless challenge for the scientific community. From the golden ratio to Birkhoff's formula, there are many mathematical models that over the centuries have been proposed to predict beauty. However, these different models have never really seduced either the artists or the philosophers, because they do not allow to account for the diversity of the aesthetic experiences, which can be produced as well by a monochrome of Klein as by a baroque canvas of Vermeer, by a real landscape or by a face.
Unlike mathematical approaches, modern artistic and philosophical theories of aesthetics are interactionist: aesthetics is a property neither of a perceived object, nor of its observer; it arises from a particular interaction between the object and the observer . Today, the challenge of scientific approaches to aesthetics is to discover and model the specificity of interactions giving rise to an aesthetic experience.
Newly inscribed in this interactionist perspective, cognitive sciences have recently proposed various biological theories making it possible to account for both the universal dimension of aesthetics and its subjective dimension. Among these theories, the aesthetic theory of fluency has rapidly established itself during the last decade, among biologists, but also artists and philosophers . According to this theory, beautiful stimuli are those whose conveyed information is processed efficiently in the brain.
The success of the theory comes in its unparalleled ability to explain both almost all of the results of experimental psychology, but also complex phenomena such as narrative suspense, which induces aesthetic pleasure by a sudden fluidification of the flow of information. The theory of fluency therefore seduces by its simplicity and its generalizing power.
However, the scientific study of fluency remains today limited by the absence of a functional definition: in psychology, fluency is only defined by the experience of ease subjectively felt by a subject. A quantitative and repeatable characterization of fluence is necessary to facilitate its scientific study, in humans and other animals.
ConvNets .– By Their performance today equal, or even superior for certain tasks, to those of humans, deep convolutional neural networks (ConvNets) represent a revolution in artificial intelligence. But recently, ConvNets have also proven to be powerful models of information processing by biological brains, surpassing the ability of previous models to predict the response of neurons in different areas of the brain .
Work in progress in our team shows a strong correlation between the sparsity of the neural code - a measure of fluency - of the different layers of a network and the attractiveness of visual stimuli (faces, paintings) evaluated empirically. These results were obtained using ConvNets pre-trained on classical object recognition tasks (eg VGG trained on ImageNet).
The objective of this internship will be twofold: - study to what extent pre-training a ConvNet with a cost function maximizing the sparsity of the neural code during training increases the correlation between the sparsity of the neural code of a test stimulus and its visual attractiveness.
- explore new measures (different from sparsity) of neuronal fluence in a ConvNet (eg, entropy).
The student should have a good level of programming in Python, basic knowledge of AI and an interest in the biological question. She / he will be immersed in the E3CO team which has several students (Master, Doctorant.es) in AI applied to biology. He / she will have access to a computer equipped with GPU cards.
Place of the internship : E3CO team, Center for Functional and Evolutionary Ecology (CEFE), Montpellier.
Date and duration of the course : adjustable according to the requirements of the course.
Gratuity : university flat rate (around 590 € / month)
In regards to Viasema was created in 2010 by a team of entrepreneurs and engineers who are experts in artificial intelligence. Jamal Rezzouk, our CTO and co-founder, graduated in Applied Mathematics from École Centrale Paris and has over 30 years of experience in the field of AI.
Viasema stands out with: Strategic projects with flagships of French industry as well as French world leaders (4 clients with more than 20 billion turnover)
The contribution of knowledge and expertise in new “state of the art” technologies that our customers do not have internally
Particularly innovative and efficient solutions Our platform is already in production in the complex IT environments of French industrial leaders demonstrating the technological maturity acquired over the past ten years
Viasema values involvement, loyalty, intelligence and rewards strong performance with increased responsibility and rapid development capabilities within the company. If you want to have an immediate impact on a fast growing company challenging the limits of intelligence and striving for excellence, we invite you to apply today.
Job description Viasema is looking for an engineer with a perfect understanding of the issues related to the implementation of Semantic Web technologies.
Beyond a diploma or an experience, we are looking for personalities, challengers, inventors at heart that we are committed to supporting in order to make your career within Viasema an enriching and stimulating experience.
You join a technical team made up of JAVA developers, experts in Artificial Intelligence (Enterprise Knowledge Graph, Semantic Web etc.) and NLP. You will be part of the Artificial Intelligence pole which is led by our AI pioneer CTO. Alongside a dynamic and motivated team, your daily life offers you a pleasant and efficient working environment where everyone can feel free to flourish through many innovative projects.
Viasema is above all a human and stimulating adventure to prepare for the future.
Your missions : · The transformation of data and data models from their current (non-semantic) form to a semantic description / modeling
· Modeling ontologies within knowledge graphs
· Management and alignment of existing ontologies to ensure interoperability between different heterogeneous sources
Required profile All profiles are likely to interest us but obviously, if you match with the following criteria, you leave with a head start:
You are a mathematics or computer engineer with in-depth knowledge of semantic web / data web, knowledge engineering / knowledge graph and Linked data.
Ability to quickly learn new technologies and familiarize themselves with new areas of activity.
The candidate must in priority master the technologies of the Semantic Web (RDF / OWL, SPARQL ...), classic ontologies such as SKOS, FRBR, Web Annotation, Schema.org ... as well as the associated tools.
Beyond technical skills, we are looking for a very autonomous and rigorous candidate, but also a good popularizer and able to be a force of proposal with technical teams as end users or management.
Guarantor of the success of your projects, you are the rare pearl if you know how to show creativity, initiative and proactivity.
Do you want to apply? Send an email to [email protected]
2-years Postdoctoral Research position at MISTEA INRAE Montpellier (https://www.inrae.fr/en/centres/occitanie-montpellier) in the context of the ANR projet DACE-DL
Areas: Semantic Web, Linked Data, Data linking
Qualifications: PhD in the domain of Semantic Web
Contact & Collaboration: Danai Symeonidou, CR INRAE Montpellier, [email protected]
Clement Jonquet, CR (HDR) INRAE Montpellier, [email protected]
Institut: INRAE is a pioneer in France in terms of data sharing and Open Science commitment. The MathNum research department gathers around 200 scientists in mathematics and digital technologies in 13 research units in France. MISTEA is a joint research unit of INRAE and Montpellier Institut Agro engineering school with activities in the development of mathematical, statistical and informatics methods dedicated to analysis and decision support for agronomy and environment. The team is also recognized for its expertise in knowledge engineering and ontology-based scientific data management and information systems.
Project context: Data linking is the scientific challenge of automatically establishing typed links between the entities of two or more structured datasets. A variety of complex data linking systems exists, evaluated on public benchmarks [1,2,3]. While they have allowed for the generation of vast amounts of linked data in the context of various dedicated projects, data generic systems often have limited applicability in many real-world scenarios, where data are highly heterogeneous and domain-specific. The ANR project DACE-DL (2022-2024) targets a paradigm shift in the data linking field with a data-centric bottom-up methodology relying on machine learning and representation learning models. We hypothesize there exists a finite number of identifiable and generalisable linking problem types (LPTs), that we need to categorize and analyse to provide better linking results.
Topic: The main goal is to identify and provide a categorisation/taxonomy of the different linking problem types based on an in-depth analysis of the linked datasets provided by the project and beyond. The first objective is to provide an in-depth analysis of the linked data available along with an exhaustive study of the state-of-the-art in the field of data linking. The postdoctoral researcher will propose a finite number of generalisable linking problem types, organised in a taxonomy encoded with OWL or SKOS, where the relations and inherent structure of the LPTs will be made explicit to both human and machine. Additionally, a (semi-)automatic taxonomy building methodology that will be easily reproducible and extensible in the future for other pairs of datasets should be built. The goal is to answer questions such as: are certain LPTs or groups of LPTs (e.g. siblings at a given level of the taxonomy) specific to a domain, language or a community? Are certain LPTs inherent to specific types of data? Once a clear and unambiguous taxonomy of LPTs is produced, various datasets will be manually annotated. These annotations on existing pairs of datasets will be used to learn, using machine learning strategies, features for the automatic categorization of other datasets.
Starting period: January 2022
Duration: 24 months
Location: INRAE Centre Occitanie-Montpellier, MISTEA, 2 Place Pierre Viala, 34000 Montpellier
Salary: Between 2200€ and 2700€ gross monthly depending on qualifications and situation.
Application: To apply for the position send an email to the contact emails ([email protected] and [email protected]):
a short description of introducing yourself
your adequacy to the position
a CV and
one major publication
The offer is also available here: https://docs.google.com/document/d/e/2PACX-1vTNvOng07Crjvwy7GbB8vKo6DhBYk0fMO3OAbQcg7pKJNxQBPE2dTKfMICX95Syyy9xLntWnIalVcJy/pub
Hello, Please find a suitable internship topic as soon as possible, in data analysis.
Do not hesitate to contact us for further information.
Eric Campo and Nadine Vigouroux
Candidacy: Send application to [email protected], [email protected] with CV, Letter motivation, Results since L1, Report of a previous internship/research dissertation.
Date: either September 2021 or January 2022 at the latest
Remuneration: Internship bonus € 591.51 / month.
Open Postdoc and PhD Positions in Computer Vision, Image Processing, and Machine Learning
(Apologies for cross-postings.) The Computer Vision Laboratory led by Prof. Radu Timofte, from the newly established Center for Artificial Intelligence and Data Science, University of Wurzburg, is looking for outstanding candidates to fill several computer vision and machine learning fully-funded postdoc and PhD positions.
Julius-Maximilians University of Würzburg (JMU), founded in 1402, is one of the leading institutions of higher education in Germany and well-known on the international stage for delivering research excellence with a global impact. The University of Würzburg is proud to be the home of outstanding researchers and fourteen Nobel Prize Laureates. Würzburg is a vibrant city in Bavaria, Germany’s economically strongest state and home base to many international companies. We look forward to welcoming you to the University of Würzburg!
Computer Vision Laboratory and University of Würzburg in general are an exciting environment for research, for independent thinking. Prof. Radu Timofte’s team is highly international, with people from about 12 countries, and the members have already won awards at top conferences (ICCV, CVPR, ICRA, NIPS, ...), founded successful spinoffs, and/or collaborated with industry. Prof. Radu Timofte is a 2022 winner of the prestigious Humboldt Professorship for Artificial Intelligence Award. Prof. Radu Timofte also leads the Augmented Perception Group at ETH Zurich.
Depending on the position, the successful candidate will focus on a subset of the following
Research Topics: • deep learning
• computational photography
• domain translation
• learned image/video compression
• image/video super-resolution
• learning paradigms
• image/video understanding
• augmented and mixed reality
• edge inference and mobile AI
The tasks will involve designing, developing, and testing novel ideas in cutting-edge research, as well as coordinating and conducting data collection for their evaluation.
The successful candidate will conduct research on deep learning machines and a new cluster with hundreds of GPUs.
The successful candidate will also collaborate with industry.
Profile • Master's degree in computer science, electrical engineering, physics or applied mathematics/ statistics
• Good programming knowledge and experience with Python / C++ / MATLAB
• Interest, prior knowledge and experience in one or more of the following is a plus: computer vision, deep learning, machine learning, image processing
• Enthusiasm for leading-edge research, team spirit, and capability of independent problem-solving
• Fluent written and spoken English is a must.
• Postdoc applicants are expected to have a strong track of published research, including top, high impact, journal (such as PAMI, IJCV, TIP, NEUCOM, JMLR, CVIU) or conference (such as ICCV, CVPR, ECCV, ICRA, NeurIPS, ICLR) papers.
Timeline The positions are open immediately, fully funded, the salaries of the doctoral students and postdocs are competitive on the German scales TV-L E13 and E14, up to 60k euros per year, before tax.
Typically a PhD takes ~4 years and a postdoc position is for a minimum of 1 year.
Applications will be reviewed on a rolling basis until all the positions are filled.
A first round of interviews is scheduled for 20.09.2021.
Application Interested applicants should email asap their PDF documents (including CV, diplomas, transcripts of records) to Prof. Radu Timofte, [email protected]
Dear Colleagues, We are looking for a postdoctoral fellow in Machine Learning applied to Biology.
Recent years have witnessed an explosion of data in biology and medicine. Many acquisition techniques, like microscopy or sequencing techniques, provide complementary views of the same system, e.g. an organ, an embryo, a tumor.
To understand the dynamics happening at single cell resolution and develop new personalized treatments, we need to integrate these complementary sources of information.
To tackle this problem, this project aims at developing new Temporal Data Integration theoretical and computational methods for various complementary acquisition techniques (microscopy, and multi-omics).
The successful candidate will work jointly within Paul Villoutreix’s interdisciplinary group (http://bioml.lis-lab.fr/) and under the supervision of Thierry Artières within the Machine Learning team of the Computer Science lab in Marseille (https://qarma.lis-lab.fr).
The position is part of the Turing Center for Living Systems (https://centuri-livingsystems.org), which is a vibrant interdisciplinary community composed of mathematicians, computer scientists, physicists, …, interested in questions of biology in the scenic Mediterranean city of Marseille, France.
We are looking for a candidate with a PhD in machine learning, computer science, applied mathematics with strong interest in machine learning and its applications to biology.
The candidate will be trained in biology and will benefit from a very strong environment of experts in machine learning. We will also provide support for applications to academic jobs and further career development.
More details in the document attached. I would be very grateful if you could spread the word to potential candidates.
Best wishes, Paul Villoutreix
We are hiring a PhD funded by Audi, on AI for autonomous driving. The post is to be filled as soon as possible.
https://karriere.audi.de/sap/bc/bsp/sap/z_hcmx_ui_ext/desktop.html#/DETAILS/CE6634AF2C021EEBBEE78DD3E0DC868A/?CSREF=NULL
In the context of the research project MultimEdia Entity Representation and Question Answering Tasks (ANR 2020-2024), a postdoctoral position is proposed for highly motivated candidates interested in multimedia understanding and natural language processing.
The main task of the post doc will consist in injecting some knowledge into a multimedia (image and text) entity representation to address Multimedia Question Answering. See the following link for details: https://www.meerqat.fr/wp-content/uploads/2021/09/meerqat_postdoc_cea_lisn.pdf
The post-doc will be supervised by CEA and LISN. The candidate will be hired by CEA (Palaiseau, near Paris, France) for a 18-months post-doc.
The LISN is located close to CEA on the Paris-Saclay University Campus.
To apply to the position, send a CV (including publication list or a URL pointing to it, such as Google Scholar) and a cover letter to Hervé Le Borgne [email protected] , Olivier Ferret [email protected] , Sahar Ghannay [email protected] and Anne Vilnat [email protected] .
You can write to [email protected] for any detail. Thank you