Approaches That Use Domain-Specific Expertise: Behavioral-Cloning-Based Advantage Actor-Critic in Basketball Games
Research on the application of artificial intelligence (AI) in games has recently gained momentum. Most commercial games still use AI based on a finite state machine (FSM) due to complexity and cost considerations. However, FSM-based AI decreases user satisfaction given that it performs the same pat...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/11/5/1110 |
| _version_ | 1827752569185763328 |
|---|---|
| author | Taehyeok Choi Kyungeun Cho Yunsick Sung |
| author_facet | Taehyeok Choi Kyungeun Cho Yunsick Sung |
| author_sort | Taehyeok Choi |
| collection | DOAJ |
| description | Research on the application of artificial intelligence (AI) in games has recently gained momentum. Most commercial games still use AI based on a finite state machine (FSM) due to complexity and cost considerations. However, FSM-based AI decreases user satisfaction given that it performs the same patterns of consecutive actions in the same situations. This necessitates a new AI approach that applies domain-specific expertise to existing reinforcement learning algorithms. We propose a behavioral-cloning-based advantage actor-critic (A2C) that improves learning performance by applying a behavioral cloning algorithm to an A2C algorithm in basketball games. The state normalization, reward function, and episode classification approaches are used with the behavioral-cloning-based A2C. The results of the comparative experiments with the traditional A2C algorithms validated the proposed method. Our proposed method using existing approaches solved the difficulty of learning in basketball games. |
| first_indexed | 2024-03-11T07:18:14Z |
| format | Article |
| id | doaj.art-16f76f60fae34a2d801de9e7184efa76 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-11T07:18:14Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-16f76f60fae34a2d801de9e7184efa762023-11-17T08:08:21ZengMDPI AGMathematics2227-73902023-02-01115111010.3390/math11051110Approaches That Use Domain-Specific Expertise: Behavioral-Cloning-Based Advantage Actor-Critic in Basketball GamesTaehyeok Choi0Kyungeun Cho1Yunsick Sung2Department of Multimedia Engineering, Dongguk University-Seoul, 30, Pildongro-1-gil, Jung-gu, Seoul 04620, Republic of KoreaDepartment of Multimedia Engineering, Dongguk University-Seoul, 30, Pildongro-1-gil, Jung-gu, Seoul 04620, Republic of KoreaDepartment of Multimedia Engineering, Dongguk University-Seoul, 30, Pildongro-1-gil, Jung-gu, Seoul 04620, Republic of KoreaResearch on the application of artificial intelligence (AI) in games has recently gained momentum. Most commercial games still use AI based on a finite state machine (FSM) due to complexity and cost considerations. However, FSM-based AI decreases user satisfaction given that it performs the same patterns of consecutive actions in the same situations. This necessitates a new AI approach that applies domain-specific expertise to existing reinforcement learning algorithms. We propose a behavioral-cloning-based advantage actor-critic (A2C) that improves learning performance by applying a behavioral cloning algorithm to an A2C algorithm in basketball games. The state normalization, reward function, and episode classification approaches are used with the behavioral-cloning-based A2C. The results of the comparative experiments with the traditional A2C algorithms validated the proposed method. Our proposed method using existing approaches solved the difficulty of learning in basketball games.https://www.mdpi.com/2227-7390/11/5/1110game AIreinforcement learningimitation learningbasketball game |
| spellingShingle | Taehyeok Choi Kyungeun Cho Yunsick Sung Approaches That Use Domain-Specific Expertise: Behavioral-Cloning-Based Advantage Actor-Critic in Basketball Games Mathematics game AI reinforcement learning imitation learning basketball game |
| title | Approaches That Use Domain-Specific Expertise: Behavioral-Cloning-Based Advantage Actor-Critic in Basketball Games |
| title_full | Approaches That Use Domain-Specific Expertise: Behavioral-Cloning-Based Advantage Actor-Critic in Basketball Games |
| title_fullStr | Approaches That Use Domain-Specific Expertise: Behavioral-Cloning-Based Advantage Actor-Critic in Basketball Games |
| title_full_unstemmed | Approaches That Use Domain-Specific Expertise: Behavioral-Cloning-Based Advantage Actor-Critic in Basketball Games |
| title_short | Approaches That Use Domain-Specific Expertise: Behavioral-Cloning-Based Advantage Actor-Critic in Basketball Games |
| title_sort | approaches that use domain specific expertise behavioral cloning based advantage actor critic in basketball games |
| topic | game AI reinforcement learning imitation learning basketball game |
| url | https://www.mdpi.com/2227-7390/11/5/1110 |
| work_keys_str_mv | AT taehyeokchoi approachesthatusedomainspecificexpertisebehavioralcloningbasedadvantageactorcriticinbasketballgames AT kyungeuncho approachesthatusedomainspecificexpertisebehavioralcloningbasedadvantageactorcriticinbasketballgames AT yunsicksung approachesthatusedomainspecificexpertisebehavioralcloningbasedadvantageactorcriticinbasketballgames |