Research map / Reinforcement learning / Visual and representation RL
Learning Diverse Policies in MOBA Games via Macro-Goals
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add sampled macro-goals and intrinsic macro-goal reward to the actor-critic policy | RL-baseline · Honor of Kings, same lineup across matches | macro-state entropy: 0.014 -> 0.408 | Helped |
| add macro-goal-guided policy learning | RL-baseline · Honor of Kings, 102 heroes vs. humans | win rate: 97.77% -> 97.81% | No clear effect |
| use MGG's macro-goal strategy | RL-baseline · Resource-Grab System | neutral monster resources snatched: +0.52 monster resources | Helped |
| use MGG's macro-goal strategy | RL-baseline · Teleportation System, Skill I | teleport rate: 0.08 -> 1.46 | Helped |
| use MGG's macro-goal strategy | RL-baseline · Teleportation System, Skill II | teleport rate: 0.40 -> 2.65 | Helped |
| use MGG instead of the common-policy RL baseline | RL-baseline · Resource-Grab System, assassin match statistics | damage: 64016.12 -> 86242.77 | Mixed |
Closest papers by meaning
- Towards Playing Full MOBA Games with Deep Reinforcement Learning (NeurIPS 2020 · similarity 0.91)
- Mastering Complex Control in MOBA Games with Deep Reinforcement Learning (AAAI 2020 · similarity 0.88)
- FightLadder: A Benchmark for Competitive Multi-Agent Reinforcement Learning (ICML 2024 · similarity 0.84)
- Learning Diverse Risk Preferences in Population-Based Self-Play (AAAI 2024 · similarity 0.84)
- Controlling Behavioral Diversity in Multi-Agent Reinforcement Learning (ICML 2024 · similarity 0.83)
- DGPO: Discovering Multiple Strategies with Diversity-Guided Policy Optimization (AAAI 2024 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.