Can I read Low-Resource Mahjong Reinforcement Learning on EtoBox?

Low-Resource Mahjong Reinforcement Learning by kmji20003 is a document available to read on EtoBox.

What is Low-Resource Mahjong Reinforcement Learning about?

The document presents the LsAc*-MJ model, a low-resource consumption reinforcement learning framework for playing Mahjong, specifically Japanese Mahjong. It utilizes long short-term memory (LSTM) neural networks and an optimized Advantage Actor-Critic (A2C) algorithm with an experience replay mechanism to enhance decision-making and training efficiency while addressing data scarcity challenges. Experimental results indicate that LsAc*-MJ outperforms traditional deep reinforcement learning models in terms of

Author

kmji20003

Language

EN