[visionlist] Two Post-doc positions in modelling the uncertainty of vision-language models

[Jose Dolz]

Two Post-doc positions in modelling the uncertainty of vision-language
models are available at ETS Montreal and CentraleSupelec Paris, starting as
soon as possible.

The main goal of this project is to create an open-sourced foundation model
for the assessment of breast cancer patients in histology images that can
adapt efficiently to novel tasks –both in terms of labeled data and
computational complexity– while yielding robust predictions, i.e., accurate
uncertainty estimates. We are expected to design and develop novel AI-based
models and algorithms for training specifically tailored for analyzing the
content of histology images and the efficient adaptation of trained models
to novel tasks, which also improve their predictive uncertainty.

Extensive experiments on real-world clinical data sets will be executed to
showcase the viability of our approach, benchmark its performance, and
analyze its advantages, limitations, and areas for improvement.

The proposed research program will devise innovative and robust algorithms
to understand the content of histology images in the context of breast
cancer (even though observations will be investigated in general computer
vision tasks too). These methods will set a new state of the art in
providing powerful foundation models that can be adapted efficiently to
novel tasks, while yielding well-calibrated predictions. In terms of
research outcomes, we expect the results from this research to be published
in the top venues of medicine (Nature Machine Intelligence), medical image
processing (MedIA, IEEE TMI, MICCAI, IPMI), computer vision (CVPR, ECCV,
ICCV) and/or machine learning (NeurIPS, ICLR, ICML). Furthermore, these
novel learning strategies will have a significant impact in the general
areas of semantic segmentation and classification in a broad span of
disciplines. In terms of clinical outcomes, the proposed methods will
provide an invaluable tool to support clinicians in the diagnosis,
treatment and follow-up of breast cancer, with a high potential to expand
it to multiple diseases. The developed algorithms will have a high impact
on many healthcare areas, such as personalized medicine and safe deployment
of AI approaches in clinical routine.



Team supervision. The École de Technologie Supérieure of Montreal (ETS) and
CentraleSupélec within the Université Paris-Saclay are opening two
postdoctoral fellowships of 18 months. The work will be conducted within an
international and dynamic environment between ETS and CentraleSupélec
locations. The potential candidates are expected to have a strong
background at the intersection of Machine Learning, Computer Vision and
Medical Imaging, who get inspired by sciences and the opportunities to
solve complex vision problems. They should have strong programming skills
and a very good understanding of data science, and Machine Learning, as
well as a strong publication track in recognized venues of computer

vision, machine learning and/or medical image computing.


*Position Qualifications*


• PhD program in Computer Science, Machine Learning, Computer Engineering,
Mathematics, or related field (e.g. applied mathematics/statistics).

• Very good understanding of Machine Learning theory and techniques, as
well as of computer vision.

• Strong publication track in recognized venues of computer vision (CVPR,
ECCV, ICCV), machine learning (NeurIPS, ICLR, ICML) and/or medical

image computing (MedIA, IEEE TMI, MICCAI).

• Good programming skills in Python (PyTorch).

• Applications/ domain-knowledge in medical image processing is a plus.

• Good communication skills in written and spoken English.

• Creativity and ability to formulate problems and solve them independently.


How to apply. Applications should be sent by email to : jose.dolz at etsmtl.ca;
pablo.piantanida at mila.quebec; maria.vakalopoulou at ecp.fr and

stergios.christodoulidis at centralesupelec.fr.


If you are interested, please send us the following elements as soon as
possible

and not later than January 31th:


• Detailed CV.

• Letter of motivation.

• Elements of bibliography or personal achievements related to research
activities.

• 2 references or recommendation letters.


-- 
*Jose Dolz*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://visionscience.com/pipermail/visionlist_visionscience.com/attachments/20240118/a45d17ee/attachment-0001.html>