From Tensors to Triplets: Redefining Deep Learning with Knowledge Graph Based Agents

In this session, we explore how to bridge the gap between deep learning and symbolic reasoning. Dr. Yoan Sallami, the creator of SynaLinks, shows how to create robust, business-ready AI systems using open graph tools.

Learn how SynaLinks's declarative, modular framework integrates seamlessly with Memgraph for real-time retrieval, schema-aware extraction, and secure LLM calls.

What’s in the Community Call?

• How SynaLinks reimagines deep learning to build self-organizing agents.
• A look at the neurosymbolic architecture behind graph-powered AI.
• A live demo of a Memgraph-integrated pipeline handling data extraction and deduplication.
• An insightful Q&A session covering topics like ambiguity and data cleanliness.

What You’ll Learn

• The fundamental differences between traditional and neurosymbolic AI approaches.
• Why a modular, Keras-style workflow simplifies the creation of complex systems.
• How to handle data consistency and schema integrity across a multi-stage pipeline.
• Expert insights on the challenges and opportunities of applying graph-powered AI.