Post-Doctorant F/H Neurosymbolic Relational Graph Neural Networks

Inria; France; Híbrido; EUR 32 000 - 45 000; Tempo integral

Resumo da oferta

Inria, France, invites applications for a Post-Doctorant in the TYREX team to design and evaluate neurosymbolic relational Graph Neural Networks for large relational databases. The project explores integrating symbolic knowledge into GraphSAGE-inspired architectures and developing a symbolic route framework with weighted and constrained messages.

You will contribute to developing a benchmark and a software prototype while preparing a rigorous experimental evaluation plan.

Descrição da oferta de emprego

Fonction : Post-Doctorant

Within the TYREX research team : tyrex.inria.fr

The objective of this postdoctoral research programme is to design, implement, and evaluate a new class of neurosymbolic relational Graph Neural Networks (GNNs) for learning over large-scale relational databases. The project focuses specifically on the integration of symbolic knowledge into scalable GNN architectures inspired by GraphSAGE and recent relational deep learning models, with a particular emphasis on the notion of atomic routes introduced in Rel-GNN. A complementary objective of the project is to contribute to the construction of a new benchmark for neurosymbolic relational learning, designed to evaluate not only predictive performance, but also the ability of models to exploit symbolic knowledge, constraints, rules, and schema-level semantics.

The first research axis will study the symbolic structures that can be extracted from relational databases and made usable by neural models. These include table schemas, primary and foreign keys, type declarations, constraints, dependencies between tuples, and domain rules. The aim is not to build a full symbolic theorem prover over relational databases, but rather to identify fragments of symbolic knowledge that can be injected as inductive biases into scalable GNN architectures. This will lead to a formal representation of symbolic annotations over relational graphs, including typed nodes, typed edges, constrained paths, rule-compatible neighborhoods, and semantically meaningful route templates.

The second research axis will focus on neurosymbolic extensions of GraphSAGE-style message passing. GraphSAGE learns inductive node representations by sampling and aggregating information from local neighborhoods. In relational databases, however, not all neighbors have the same semantic role. Some links encode entity ownership, some encode transactions, some encode temporal events, and others encode many-to-many relationships through bridge tables. The project will therefore investigate symbolic aggregation mechanisms in which messages are filtered, weighted, or transformed according to schema-level and rule-level information. For example, aggregation functions may be conditioned by relation types, constrained by integrity rules, or regularized so that learned embeddings respect known dependencies.

The third and most specific research axis will build upon Rel-GNN’s notion of atomic routes. Atomic routes provide elementary relational paths that support structured information exchange between source and destination nodes, especially in the presence of bridge tables or higher-order relational patterns. This project will study how atomic routes can be made neurosymbolic. Rather than considering atomic routes only as structural units for efficient message passing, we will enrich them with symbolic semantics. A route may be typed, associated with a logical rule, constrained by a dependency, or assigned an interpretation derived from the database schema. The model will then learn not only which routes are predictive, but also how symbolic constraints should guide route selection and composition.

Several technical directions will be explored. One direction is symbolic route filtering, where only routes compatible with schema constraints or expert rules are made available to the GNN. Another direction is differentiable symbolic route weighting, where rules provide soft priors over message passing paths. A third direction is logic-aware regularization, where the loss function penalizes predictions or intermediate representations that violate known constraints. A fourth direction is interpretable route attribution, where predictions can be explained in terms of the atomic routes and symbolic rules that contributed most strongly to the output.

A further axis of the postdoctoral programme will be devoted to the construction of a benchmark for neurosymbolic relational learning. Current benchmarks for relational learning provide important datasets and predictive tasks, but they are not primarily designed to assess the contribution of symbolic knowledge. The project will therefore contribute to the definition of benchmark tasks where schemas, constraints, dependencies, rules, and expert knowledge are explicitly represented and can be used by learning systems. The benchmark will aim to evaluate several dimensions: predictive accuracy, data efficiency, robustness to missing or noisy data, generalization under distribution shift, constraint satisfaction, and interpretability of relational reasoning paths. The postdoctoral researcher will participate in the design of benchmark tasks, the preparation of datasets, the formalization of symbolic annotations, and the implementation of evaluation protocols.

The benchmark will also serve as an experimental testbed for the neurosymbolic GNN architectures developed in the project. In particular, it will make it possible to compare purely statistical relational GNNs with models that exploit symbolic route filtering, rule-guided aggregation, constraint-aware training, or logic-aware route attribution. This will allow the project to move beyond standard predictive evaluation and assess whether symbolic knowledge improves learning in settings where it should be useful: low-label regimes, rare relational patterns, schema-rich databases, noisy relations, unseen entity types, and tasks requiring consistency with domain rules.

The postdoctoral work will include the development of a software prototype compatible with existing relational learning pipelines. The prototype will implement neurosymbolic variants of GraphSAGE and Rel-GNN-style message passing, support automatic extraction of symbolic metadata from relational schemas, and provide experimental modules for route filtering, rule-guided aggregation, constraint-aware training, and benchmark evaluation. The implementation will be designed to remain scalable on large relational datasets and to support systematic comparison with existing relational GNN baselines.

The experimental evaluation will be conducted on relational learning benchmarks such as RelBench, on the newly developed neurosymbolic relational learning benchmark, and, if available, on real relational databases from scientific or industrial domains. The evaluation will address predictive performance, data efficiency, robustness to missing or noisy data, generalization under distribution shift, computational cost, constraint satisfaction, and interpretability. Particular attention will be given to tasks where domain rules, schema constraints, or route-level symbolic information are expected to provide a measurable advantage.

Avantages

Applications must be submitted online via the Inria website. Processing of applications submitted via other channels is not guaranteed.