发表机构
Sun Yat-sen University; North China Electric Power University; Unilumin Group Co., Ltd.（中山大学; 华北电力大学; 洲明集团）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
CompEvo提出竞争诱导进化框架，通过进化博弈理论保证收敛，使多智能体在新闻驱动时间序列预测中保持多样性，实验平均降低RMSE 27.3%和MAPE 26.2%。
AI 中文摘要
新闻驱动的时间序列预测利用不断演变的文本事件以及历史观测值来预测未来数值，支撑市场风险监测和资源调度等应用。在多智能体设置中，仍存在两个挑战。第一个是思维退化，即智能体收敛到相似的证据寻求行为。第二个是理论依据不足，即策略更新往往是启发式的，缺乏原则性的公式化表述。为应对上述挑战，我们提出了CompEvo，一个用于多智能体新闻驱动时间序列预测的竞争诱导进化框架。在理论依据方面，我们引入了一个进化博弈公式，以保证均衡存在性和优化收敛性。基于该公式，我们构建了一个可训练的多智能体进化框架，该框架整合了策略执行、基于适应度的可微选择以及竞争诱导的策略进化。CompEvo使异构智能体能够探索多样的新闻证据，将预测反馈转换为可微的影响权重，并在竞争压力下进化智能体策略，以在保持多样性的同时保留有效逻辑。在四个真实世界数据集上的实验表明，与强基线相比，CompEvo平均将RMSE降低了27.3%，MAPE降低了26.2%。进一步的分析表明，CompEvo成功地维持了多样化和专业化的智能体行为。
英文摘要
News-driven time series forecasting uses evolving textual events together with historical observations to predict future values, supporting applications such as market risk monitoring and resource scheduling. In multi-agent settings, two challenges still remain. The first is degeneration of thought, where agents converge to similar evidence-seeking behaviors. The second is insufficient theoretical grounding, where strategy updates are often heuristic and lack a principled formulation. To address the above challenges, we propose CompEvo, a competition-induced evolution framework for multi-agent news-driven time series forecasting. For theoretical grounding, we introduce an evolutionary game formulation to guarantee equilibrium existence and optimization convergence. Building on this formulation, we construct a trainable multi-agent evolution framework that integrates strategy execution, fitness-based differentiable selection, and competition-induced strategy evolution. CompEvo enables heterogeneous agents to explore diverse news evidence, converts forecasting feedback into differentiable influence weights, and evolves agent strategies under competitive pressure to preserve effective logic while maintaining diversity. Experiments on four real-world datasets show that CompEvo reduces RMSE by 27.3% and MAPE by 26.2% on average over strong baselines. Further analysis indicates that CompEvo successfully maintains diverse and specialized agent behaviors.
