Robust Satisficing Ensemble：分布偏移下可扩展的模型聚合新框架
Robust Satisficing Ensemble: Scalable Model Aggregation Under Distribution Shifts
集成模型在训练分布外容易翻车，这篇论文的 RSE 把脆弱性最小化变成了可解析求解的问题，跑分上压过了 GroupDRO 和 IRM 这些老牌鲁棒方法。做模型部署或鲁棒性研究的可以看看推导，工程上也能直接落地。
arXiv 论文提出 Robust Satisficing Ensemble（RSE），一种面向分布偏移场景的概率化模型聚合框架。
其核心机制是将鲁棒满足目标与二次风险替代项结合，推导出解析形式的内层对手，从而绕开标准 minimax 优化的计算瓶颈，使优化复杂度与样本量解耦，可扩展到大规模数据集。理论上，论文通过 Lagrange-Burmann 展开证明脆弱性近似的截断误差按几何速率衰减，并给出 RSE 的驻点保证。
实验覆盖对抗性标签偏移（SST-5、TREC）、自然子群偏移与时间偏移（CivilComments、HuffPost）等场景，RSE 持续优于 GroupDRO、IRM、SRM 等强鲁棒基线。对需要在新数据分布上保持稳定表现的工程团队，这提供了一种有理论支撑且计算可行的集成方案。
Robust Satisficing Ensemble: Scalable Model Aggregation Under Distribution Shifts
Standard ensemble methods can improve robustness by mitigating individual model stochasticity, but they frequently fail to generalize beyond the training distribution, leaving systems vulnerable to out-of-distribution (OOD) data. We introduce the Robust Satisficing Ensemble (RSE), a scalable probabilistic aggregation framework designed to secure model combinations by minimizing their fragility against distributional changes. By bridging a robust satisficing objective with a quadratic risk surrogate, RSE derives an analytical inner adversary. This fundamental reformulation enables highly efficient weight updates without the computational bottleneck of standard minimax optimization, crucially decoupling optimization complexity from sample size to scale effectively to large datasets. In particular, we prove a geometrically decaying truncation error for the fragility approximation via Lagrange-Burmann expansion and establish stationarity guarantees for RSE. We evaluate RSE across a comprehensive spectrum of distribution shifts, including adversarial label shift (SST-5, TREC), natural subpopulation and temporal shifts (CivilComments, HuffPost). Our results demonstrate that RSE consistently outperforms strong robust baselines, including GroupDRO, IRM, and SRM. These findings establish RSE as a practical, theoretically grounded safeguard for reliable model aggregation in dynamic environments.
This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include RSE, 分布偏移, 模型集成, GroupDRO.
Source attribution: curated from arXiv cs.LG (@Ahmet Faruk Cetinkaya, Enes Agirman, Cem Tekin).
