RaidEnv: Exploring New Challenges in Automated Content Balancing for Boss Raid Games

• DOI: 10.48550/arxiv.2307.01676
• Publication Date: 2023-01-01
• Authors: Hyeonchang Jeon; In-Chang Baek; Cheong-mok Bae; Taehwa Park; Wonsang You; Tae-Gwan Ha; Hoyun Jung; Jin-Ha Noh; Seung‐Won Oh; Kyung-Joong Kim
• Keywords: Boss; RAID; Content (measure theory)

The balance of game content significantly impacts the gaming experience. Unbalanced diminishes engagement or increases frustration because repetitive failure. Although designers intend to adjust difficulty content, this is a repetitive, labor-intensive, and challenging process, especially for commercial-level games with extensive content. To address issue, research community has explored automated balancing using artificial intelligence (AI) techniques. However, previous studies have focused on limited did not consider importance generalization ability playtesting agents when encountering changes. In study, we propose RaidEnv, new simulator that includes diverse customizable boss raid scenario in MMORPG games. Additionally, design two benchmarks can aid practical application AI. These open problems automatic balancing, introduce evaluation metrics provide guidance AI balancing. This novel platform expands frontiers offers framework within realistic production pipeline.