arXiv：cs.LG· Jianxiang Xie, Belal Alsinglawi·· 4 小时前AI 评分36
面向可靠选择性预测的结构感知图弃权方法
Structure-Aware Graph Abstention for Reliable Selective Forecasting
AI 导读
该研究提出结构感知图弃权方法，将实例级合理性与关系一致性视为两条独立可靠性轴，通过可学习稀疏图和 Dirichlet 式结构能量 E_struct 实现关系一致性打分，并用误差加权图正则化与分数误差对齐训练。在七个长时程基准和四个骨干网络上，结构门控在相同覆盖率下常能降低选择性 MSE，优于 TEM，增益在跨变量结构信息更丰富的场景最大。增益并非普遍存在，表明其为互补的弃权信号。
正文
Abstract:Selective forecasting abstains on high-risk test windows under a retained-coverage budget. Existing gates such as TEM (Brusokas et al., 2025) score each forecast as a whole; for multivariate outputs, trajectories can look plausible while violating dependencies among variables. We treat instance-level plausibility and relational consistency as distinct reliability axes and operationalize the latter via a learned sparse graph and a Dirichlet-style structural energy E_struct, trained with error-weighted graph regularization and score-error alignment. On seven long-horizon benchmarks and four backbones, structural gating often reduces selective MSE versus TEM at matched coverage, with the largest gains where cross-variable structure appears more informative in our benchmarks; gains are not universal, indicating a complementary abstention signal. Table 1 is a Protocol A ranking diagnostic (seed 2024); three-seed deployable Protocol B on an aligned subset is in Table 3 (full validation-to-test grids: Appendix A).
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.08322 [cs.LG] |
| (or arXiv:2610.08322v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.08322 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jianxiang Xie [view email]
[v1]
Tue, 6 Oct 2026 13:23:51 UTC (1,505 KB)
来源：arXiv：cs.LG · arxiv.org