Dear Marmamers,

We are now recruiting a postdoc in Machine Learning for beluga re-identification in the St. Lawrence Estuary. The selected candidate will have the opportunity to work on a huge database of more than 30 years of photo-ID! All details below.

Do not hesitate to contact me if you have any questions.

Best regards,

Clément

Project title

Machine learning approach for the automatic re-identification of individuals of the St. Lawrence Estuary Beluga whales population using photo-identification data.

Context and Objective

Re-identifying animals in the wild using photo-identification (photo-ID) allows to determine key ecological parameters of wildlife populations (e.g., abundance, carrying capacity, community structure). For over 31 years, the Group for Research and Education on Marine Mammals (GREMM) has built and maintained a photo-ID database for the population of the St. Lawrence Estuary Beluga (SLEB) taken from a research boat. Those shots of belugas’ flanks allowed to identify about 350 individuals among the whole population, which counts around 1000 individuals. The traditional re-identification process involves the examination of each photo by an operator to locate any distinctive marks allowing to either identify a known individual or add a new one to the database. This task is a very tedious task, and in the past 31 years, only 21 years of photo-ID have been partially processed. Moreover, more photos are taken every year than it is possible to process manually.

However, those data are of major importance to characterize the social dynamics of the SLEB population. Indeed, the recognition of individuals making different herds is the key to identify distinct communities, which each have their own pattern of use of the summer habitat and site-fidelity. Therefore, the vulnerability of each SLEB community with regard to human activities (e.g., navigation) depends on spatiotemporal patterns of habitat use that must be characterized to the best of our knowledge. In this context, the objective of the project is to investigate a machine learning approach (e.g., based on Siamese neural  networks), but also the existing relevant literature from the privacy domain, for the automatic re-identification of known individuals from the beluga population but also to discover new individuals that were unregistered until then. The automatization of this process will notably enable the reduction of the biases introduced by the human judgement for the recognition of the individuals with distinctive marks.

The postdoctoral candidate must have the following qualities: motivation, curiosity, sense of initiative, autonomy, creativity as well as demonstrating excellent capacities to work as part of a team. He/she will be encouraged to travel to present the results to international scientific conferences, in addition to the meeting with partners and relevant stakeholders. The selected candidate will also contribute to the production and writing of project deliverables.

Background and skills

●      PhD in Computer Science and/or recognized experience in the development of unsupervised and supervised machine learning algorithms

●      Experience in technology transfer

●      Capacity to write grant and funding demands

●      Ability to communicate (oral and written) scientific results to experts and non-experts, including the writing of scientific papers and the review of the state-of-the-art in English

Assets

●      Experience in Siamese neural networks

●      Experience in deep learning and/or in anonymisation/re-identification

●      Proficiency in spoken French

Period

●      Start: Now

●      End: March 31, 2022

Rémunération

●      50 k$/year + funding available for conferences

Location

●      Gatineau, Montréal or Ripon (Québec, Canada)

Application procedure

●      Send by email a CV (academic format), cover letter outlining your motivation along with your skills and assets with regard to the project, and the name and contact information of three referees from Academia :

○      Pr Clément Chion (clement.chion@uqo.ca)

○      Pr Sébastien Gambs (gambs.sebastien@uqam.ca)

●      Deadline : July 31, 2020, or until the position is filled