Overview
This postdoctoral position offers a unique opportunity to delve into the applications of Optimal Transport (OT) in the analysis of graph data. Located at the Institute 3IA Côte d’Azur in Nice, France, this role is significant for advancing research in machine learning and artificial intelligence, particularly in the context of graph clustering and generative models.
Background & Relevance
Optimal Transport has emerged as a crucial tool in machine learning, enabling effective training of models by providing meaningful gradients on empirical distributions. The Wasserstein distance, a key concept in OT, facilitates the encoding of geometrical information, enhancing model interpretability. The focus of this position is on the Gromov-Wasserstein (GW) transport distance, which has shown promise in measuring similarity between labeled graphs. The Fused Gromov-Wasserstein (FGW) distance variant has been identified as a robust measure for graph classification and clustering, making this research area particularly relevant in the current landscape of machine learning.
Key Details
- Location: Nice, France, at the Lagrange Laboratory, Parc Valrose.
- Duration: Initial funding for 1 year, extendable to 2 years based on candidate performance.
- Start Date: As early as October/November 2019.
- Teaching Requirement: Expected to teach 64 hours in Machine Learning/Data Science.
- Application Links: Postdoc Offer
Eligibility & Participation
Candidates with a strong mathematical and optimization background, along with experience in machine learning or graph processing, are encouraged to apply. Proficiency in Python is essential for this role.
Submission or Application Guidelines
Interested candidates should submit their CV to the following email addresses:
– remi.flamary@unice.fr
– charles.bouveyron@math.cnrs.fr
Additional Context / Real-World Relevance
The research conducted in this postdoctoral position will not only contribute to theoretical advancements in graph analysis but also have practical implications in various domains, such as social and communication networks. Understanding graph structures through optimal transport methods can lead to improved clustering techniques and better insights into complex data relationships.
Conclusion
This postdoctoral opportunity represents a significant step for researchers looking to advance their careers in machine learning and graph analysis. Interested individuals are encouraged to apply and contribute to this innovative research area.
Category: PhD & Postdoc Positions
Tags: optimal transport, graph data, machine learning, postdoc, cote d’azur, gromov-wasserstein, data science, graph clustering
