Scholar
Mikayel Samvelyan
Google Scholar ID: 2Qs19WAAAAAJ
Google DeepMind
Citations & Impact
Citations
16,286
H-index
14
i10-index
17
Publications
20
Co-authors
56
list available
Publications
Resume (English only)
Academic Achievements
- Developed widely-used multi-agent RL tools: QMIX method and SMAC benchmark.
- Released SMACv2 (accepted at NeurIPS 2023) and MADRID (accepted at AAMAS 2024).
- MAESTRO accepted at ICLR 2023.
- Proposed Rainbow Teaming (NeurIPS 2024), an approach for generating diverse adversarial prompts to improve LLM robustness.
- Contributed to the release of Meta Llama 3 (2024), the most capable openly available LLM to date.
- Developed JaxMARL (accepted at NeurIPS 2024).
- Co-organized the 2nd Workshop on Agent Learning in Open-Endedness (ALOE) at NeurIPS 2023.
- Delivered a keynote on 'Agent Learning in Open-Endedness' at NeurIPS 2024.
