Research map / Reinforcement learning / Policy gradient and actor-critic
Stackelberg Actor-Critic: Game-Theoretic Reinforcement Learning Algorithms
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 |
|---|---|---|---|
| Use Stackelberg actor-leader updates instead of individual-gradient updates | Actor-critic algorithms · OpenAI Gym MuJoCo tasks | average episode return | Helped |
| Use STDDPG-AL instead of DDPG | DDPG · OpenAI Gym MuJoCo tasks | average episode return | Helped |
| Use STDDPG-CL instead of DDPG | DDPG · OpenAI Gym MuJoCo tasks | average episode return | Mixed |
| Use STSAC-AL instead of SAC | SAC · OpenAI Gym MuJoCo tasks | average episode return | No clear effect |
| Use the actor as leader rather than the critic | Stackelberg actor-critic algorithms · OpenAI Gym MuJoCo tasks | average episode return | Helped |
| Use multiple critic unrolling steps (ablation) | STAC · Actor-critic tasks | average episode return | Helped |
| Use STAC with m=1 instead of AC with m=80 (ablation) | STAC and AC · CartPole | average episode return | Helped |
| Add entropy regularization to the actor objective (ablation) | Individual-gradient and Stackelberg-gradient updates · Motivating example | convergence speed | No clear effect |
| Add entropy regularization to the actor objective (ablation) | Individual-gradient and Stackelberg-gradient updates · Motivating example | performance gap between updates | Hurt |
Closest papers by meaning
- Characterizing the Gap Between Actor-Critic and Policy Gradient (ICML 2021 · similarity 0.92)
- Bi-Level Actor-Critic for Multi-Agent Coordination (AAAI 2020 · similarity 0.92)
- Learning in Stackelberg Mean Field Games: A Non-Asymptotic Analysis (NeurIPS 2025 · similarity 0.92)
- Oracles & Followers: Stackelberg Equilibria in Deep Multi-Agent Reinforcement Learning (ICML 2023 · similarity 0.90)
- Finite-Time Global Optimality Convergence in Deep Neural Actor-Critic Methods for Decentralized Multi-Agent Reinforcement Learning (ICML 2025 · similarity 0.87)
- Structured Reinforcement Learning for Combinatorial Decision-Making (NeurIPS 2025 · similarity 0.87)
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.