Open AccessEvaluating the impact of curriculum learning on the training process for an intelligent agent in a video game
Author(s) -
Jorge E. Camargo,
Rigoberto Sáenz
Publication year - 2021
Publication title -
inteligencia artificial
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.149
H-Index - 12
eISSN - 1988-3064
pISSN - 1137-3601
DOI - 10.4114/intartif.vol24iss68pp1-20
Subject(s) - curriculum , reinforcement learning , computer science , video game , process (computing) , measure (data warehouse) , training (meteorology) , multimedia , artificial intelligence , mathematics education , psychology , pedagogy , data mining , physics , meteorology , operating system
We want to measure the impact of the curriculum learning technique on a reinforcement training setup, several experiments were designed with different training curriculums adapted for the video game chosen as a case study. Then all were executed on a selected game simulation platform, using two reinforcement learning algorithms, and using the mean cumulative reward as a performance measure. Results suggest that curriculum learning has a significant impact on the training process, increasing training times in some cases, and decreasing them up to 40% percent in some other cases.