Learning to Play Blackjack: A Curriculum Learning Perspective

TL;DR

This paper introduces a curriculum learning framework using Large Language Models to improve reinforcement learning agents in Blackjack, resulting in faster training and better performance.

Contribution

It presents a novel LLM-guided curriculum approach for RL that enhances learning efficiency and effectiveness in complex environments like Blackjack.

Findings

DQN agent's win rate increased from 43.97% to 47.41%.

Bust rate reduced from 32.9% to 28.0%.

Training workflow accelerated by over 74%.

Abstract

Reinforcement Learning (RL) agents often struggle with efficiency and performance in complex environments. We propose a novel framework that uses a Large Language Model (LLM) to dynamically generate a curriculum over available actions, enabling the agent to incorporate each action individually. We apply this framework to the game of Blackjack, where the LLM creates a multi-stage training path that progressively introduces complex actions to a Tabular Q-Learning and a Deep Q-Network (DQN) agent. Our evaluation in a realistic 8-deck simulation over 10 independent runs demonstrates significant performance gains over standard training methods. The curriculum-based approach increases the DQN agent's average win rate from 43.97% to 47.41%, reduces the average bust rate from 32.9% to 28.0%, and accelerates the overall workflow by over 74%, with the agent's full training completing faster than the baseline's evaluation phase alone. These results validate that LLM-guided curricula can build more effective, robust, and efficient RL agents.