This repository contains graph-based data models for implementing different types of agent memory systems. These models are designed to represent and manage various aspects of agent memory in a structured, graph-based format.
Agent memory is a crucial component in AI systems, enabling agents to maintain context, learn from past experiences, and make informed decisions. This repository provides graph-based implementations of different memory paradigms, allowing for flexible and scalable memory management in agent systems.
The repository contains implementations of various memory models, each represented as a graph structure. These models can be used independently or combined to create more complex memory systems.
The repository includes implementations for:
- Episodic Memory: Captures and stores specific events and experiences
- Semantic Memory: Manages general knowledge and facts
- Procedural Memory: Stores learned skills and procedures
- Temporal Memory: Handles time sensitive information processing
