AI Metropolis: Scaling Large Language Model-based Multi-Agent Simulation with Out-of-order Execution
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916470090366976 |
|---|---|
| author | Xie, Zhiqiang Kang, Hao Sheng, Ying Krishna, Tushar Fatahalian, Kayvon Kozyrakis, Christos |
| author_facet | Xie, Zhiqiang Kang, Hao Sheng, Ying Krishna, Tushar Fatahalian, Kayvon Kozyrakis, Christos |
| contents | With more advanced natural language understanding and reasoning capabilities, large language model (LLM)-powered agents are increasingly developed in simulated environments to perform complex tasks, interact with other agents, and exhibit emergent behaviors relevant to social science and gaming. However, current multi-agent simulations frequently suffer from inefficiencies due to the limited parallelism caused by false dependencies, resulting in performance bottlenecks. In this paper, we introduce AI Metropolis, a simulation engine that improves the efficiency of LLM agent simulations by incorporating out-of-order execution scheduling. By dynamically tracking real dependencies between agents, AI Metropolis minimizes false dependencies, enhancing parallelism and enabling efficient hardware utilization. Our evaluations demonstrate that AI Metropolis achieves speedups from 1.3x to 4.15x over standard parallel simulation with global synchronization, approaching optimal performance as the number of agents increases. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_03519 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | AI Metropolis: Scaling Large Language Model-based Multi-Agent Simulation with Out-of-order Execution Xie, Zhiqiang Kang, Hao Sheng, Ying Krishna, Tushar Fatahalian, Kayvon Kozyrakis, Christos Distributed, Parallel, and Cluster Computing Artificial Intelligence Machine Learning Multiagent Systems With more advanced natural language understanding and reasoning capabilities, large language model (LLM)-powered agents are increasingly developed in simulated environments to perform complex tasks, interact with other agents, and exhibit emergent behaviors relevant to social science and gaming. However, current multi-agent simulations frequently suffer from inefficiencies due to the limited parallelism caused by false dependencies, resulting in performance bottlenecks. In this paper, we introduce AI Metropolis, a simulation engine that improves the efficiency of LLM agent simulations by incorporating out-of-order execution scheduling. By dynamically tracking real dependencies between agents, AI Metropolis minimizes false dependencies, enhancing parallelism and enabling efficient hardware utilization. Our evaluations demonstrate that AI Metropolis achieves speedups from 1.3x to 4.15x over standard parallel simulation with global synchronization, approaching optimal performance as the number of agents increases. |
| title | AI Metropolis: Scaling Large Language Model-based Multi-Agent Simulation with Out-of-order Execution |
| topic | Distributed, Parallel, and Cluster Computing Artificial Intelligence Machine Learning Multiagent Systems |
| url | https://arxiv.org/abs/2411.03519 |