Research map / Multi-agent systems and games / Games and equilibria
Strategic Knowledge Transfer
Max Olan Smith, Thomas Anthony, Michael P. Wellman
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| transfer pure-strategy best responses by Q-Mixing instead of training BR directly on the mixture | Double DQN LSTM policy · Running-with-Scissors, uniform mixed opponent | return | Improved |
| use Q-Mixing instead of BR(Uniform) across the opponent mixture space | Double DQN LSTM policy · Running-with-Scissors, all mixtures in tenths | return and normalized return | Improved |
| average Q-values of more independently trained DQNs (ablation) | DQN · Running-with-Scissors, uniform mixed opponent | return: 12 of the 25 combinations improved | No clear effect |
| add an opponent classifier (OPC) to Q-Mixing | Q-Mixing: Prior · Running-with-Scissors, opponent mixture coverage | return | Improved |
| use Q-Mixing: Prior instead of pure-strategy best responses | Best responses to pure opponent policies · Running-with-Scissors, opponent mixtures | normalized return | Improved |
| use observation-frequency likelihood (Freq) instead of prior only | Q-Mixing: Prior · Running-with-Scissors | return | Hurt |
| use Q-Mixing: Uniform Prior instead of BR(Uniform) | Q-Mixing · Running-with-Scissors | return | Improved |
| add information-gathering response via Q-Mixing in observe-or-pass game | Q-Mixing (single-step) · Opponent-policy identification game | return: 0.97 -> 0.5 | Hurt |
Closest papers by meaning
- The Sample Complexity of Online Strategic Decision Making with Information Asymmetry and Knowledge Transportability (ICML 2025 · similarity 0.88)
- Is Knowledge Power? On the (Im)possibility of Learning from Strategic Interactions (NeurIPS 2024 · similarity 0.87)
- A-PSRO: A Unified Strategy Learning Method with Advantage Metric for Normal-form Games (ICML 2025 · similarity 0.87)
- Scaling Inference-Time Computation via Opponent Simulation: Enabling Online Strategic Adaptation in Repeated Negotiation (ICML 2026 · similarity 0.86)
- LOQA: Learning with Opponent Q-Learning Awareness (ICLR 2024 · similarity 0.86)
- Mimicking To Dominate: Imitation Learning Strategies for Success in Multiagent Games (NeurIPS 2024 · similarity 0.86)
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.
