Postdoc: Neurosymbolic Relational GNNs - Flexible Hours

inria; France; On-site; EUR 30,000 - 32,000; Full time

Benefits offered by this job

Transport reimbursement

Annual leave

Teleworking after 6 months

Equipment provided

Social events

Training access

Social security coverage

Job summary

Inria's TYREX research team in France invites applications for a postdoctoral researcher on symbolic knowledge integration into relational GNNs. The project aims to guide message passing in graph models using schemas and rules, building neurosymbolic systems that improve learning from relational data.

The successful candidate will develop principled methods, implement experiments, and contribute to benchmarks and publications.

Qualifications

• PhD in computer science or related field.
• Strong background in graph neural networks.
• Experience with neurosymbolic methods is a plus.

Responsibilities

• Design and evaluate neurosymbolic relational learning models.
• Publish results in AI and databases venues.
• Collaborate with the TYREX team and Inria researchers.

Skills

Relational GNNs

Neurosymbolic ML

Graph Neural Nets

Python

ML research

Education

PhD in Computer Science

Tools

PyTorch

NetworkX

NumPy

Job description

Inria's TYREX research team in France invites applications for a postdoctoral researcher on symbolic knowledge integration into relational GNNs. The project aims to guide message passing in graph models using schemas and rules, building neurosymbolic systems that improve learning from relational data.

The successful candidate will develop principled methods, implement experiments, and contribute to benchmarks and publications.