Li, HualeCao, RuiHou, XiaohanWang, XuanTang, LinlinZhang, JiajiaQi, ShuhanIn recent years, deep reinforcement learning achieves great success in many fields, especially in the field of games, such as AlphaGo, AlphaZero and AlphaStar. However, reward sparsity is still a problem in the 3D strategy games with a higher dimension of state space and more complex game scenarios. To solve this problem, in this paper, we propose an intrinsic-based policy optimization algorithm (IBPO) for reward sparsity. The IBPO incorporates the intrinsic reward into the traditional policy, which composed by the differential fusion mechanism and the modified value network. The experimental results show our method can obtain better performance than the previous methods on the VizDoom.
IBPO: Solving 3D Strategy Game with the Intrinsic Reward
Smart Innovation, Systems and Technologies
Advances in Smart Vehicular Technology, Transportation, Communication and Applications ; Chapter : 25 ; 257-264
2021-11-30
8 pages
Article/Chapter (Book)
Electronic Resource
English
IBPO: Solving 3D Strategy Game with the Intrinsic Reward
TIBKAT | 2022
|REWARD POINT MANAGEMENT DEVICE, REWARD POINT MANAGEMENT METHOD, AND REWARD POINT MANAGEMENT SYSTEM
European Patent Office | 2020
|REWARD ISSUANCE DEVICE, REWARD ISSUANCE SYSTEM, VEHICLE-MOUNTED DEVICE, AND PROGRAM
European Patent Office | 2020
|