mlsanityData to 7 Oct 2026

Learning Diverse Policies in MOBA Games via Macro-Goals

arXiv 2110.14221 ↗First posted 27 Oct 2021cs.LG cs.AIcited by 7

Not on the map. Why?

Abstract (arXiv)

Recently, many researchers have made successful progress in building the AI systems for MOBA-game-playing with deep reinforcement learning, such as on Dota 2 and Honor of Kings. Even though these AI systems have achieved or even exceeded human-level performance, they still suffer from the lack of policy diversity. In this paper, we propose a novel Macro-Goals Guided framework, called MGG, to learn diverse policies in MOBA games. MGG abstracts strategies as macro-goals from human demonstrations and trains a Meta-Controller to predict these macro-goals. To enhance policy diversity, MGG samples macro-goals from the Meta-Controller prediction and guides the training process towards these goals. Experimental results on the typical MOBA game Honor of Kings demonstrate that MGG can execute diverse policies in different matches and lineups, and also outperform the state-of-the-art methods over 102 heroes.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Builds on 23 of its 30 references that are in the map. First posted on arXiv 27 Oct 2021.

Builds on

23 of 30 references.

References · 23

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2017Deep reinforcement learning from human preferencescited by 3,846
  3. 2016Generative Adversarial Imitation Learningcited by 1,576
  4. 2017Hindsight Experience Replaycited by 1,096
  5. 2019Dota 2 with Large Scale Deep Reinforcement Learningcited by 1,084
  6. 2018Data-Efficient Hierarchical Reinforcement Learningcited by 436
  7. 2017Population Based Training of Neural Networkscited by 351
  8. 2017Overcoming Exploration in Reinforcement Learning with Demonstrationscited by 322
  9. 2018Reward learning from human preferences and demonstrations in Ataricited by 314
  10. 2016Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivationcited by 282
  11. 2018Human-level performance in first-person multiplayer games with population-based deep reinforcement learningcited by 260
  12. 2017Automatic Goal Generation for Reinforcement Learning Agentscited by 252
11 more
  1. 2017DeepStack: Expert-Level Artificial Intelligence in No-Limit Pokercited by 228
  2. 2019Mastering Complex Control in MOBA Games with Deep Reinforcement Learningcited by 150
  3. 2018Deep Counterfactual Regret Minimizationcited by 102
  4. 2019Open-ended Learning in Symmetric Zero-sum Gamescited by 96
  5. 2020Towards Playing Full MOBA Games with Deep Reinforcement Learningcited by 78
  6. 2020Supervised Learning Achieves Human-Level Performance in MOBA Games: A Case Study of Honor of Kingscited by 18
  7. 2021MapGo: Model-Assisted Policy Optimization for Goal-Oriented Taskscited by 13
  8. 2018Hierarchical Macro Strategy Model for MOBA Game AIcited by 12
  9. 2020Which Heroes to Pick? Learning to Draft in MOBA Games with Neural Networks and Tree Searchcited by 10
  10. 2021Boosting Offline Reinforcement Learning with Residual Generative Modelingcited by 7
  11. 2017MOBA: a New Arena for Game AIcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 3; 2025: 1; 2026: 1.

7 papers.

  1. 2023TiZero: Mastering Multi-Agent Football with Curriculum Learning and Self-Playcited by 18
  2. 2024Hokoff: Real Game Dataset from Honor of Kings and its Offline Reinforcement Learning Benchmarkscited by 10
  3. 2023Towards Effective and Interpretable Human-Agent Collaboration in MOBA Games: A Communication Perspectivecited by 5
  4. 2024Enhancing Human Experience in Human-Agent Collaboration: A Human-Centered Modeling Approach Based on Positive Human Gaincited by 4
  5. 2025RIFT: Group-Relative RL Fine-Tuning for Realistic and Controllable Traffic Simulationcited by 4
  6. 2026CRAFT: Counterfactual-to-Interactive Reinforcement Fine-Tuning for Driving Policiescited by 2
  7. 2024Offline Imitation of Badminton Player Behavior via Experiential Contexts and Brownian Motioncited by 0