DeepYardDeepYard
A

AgenticRec

Agentic framework for optimizing recommendation system rankings through end-to-end policy learning

Open SourceFree

About

AgenticRec is a research framework that applies agentic AI principles to recommender systems, optimizing the entire decision-making pipeline from reasoning to tool usage to final item ranking. Uses end-to-end policy optimization to improve recommendation quality by treating the recommendation process as a sequential decision task. Designed for researchers exploring the intersection of reinforcement learning, LLM agents, and recommendation systems.

Details

Language
Patterns

Tags

frameworkautonomousopen-sourcepythontool-useorchestration