arXiv：cs.LG· Louis Berthier, Ahmed Shokry, Maxime Moreaud, Guillaume Ramelet, Aymeric Dieuleveut·· 16 小时前
自组织共形预测 SOCP：用无监督分组缩小区域覆盖缺口
Self-Organized Conformal Prediction: Reducing Regional Coverage Gaps with Unsupervised Group Discovery
AI 导读
研究者提出自组织共形预测（SOCP），用无需校准数据的自组织映射（SOM）发现输入空间分组，预测时由查询的最佳匹配单元（BMU）从单个 cell、固定网格邻域或原型扩展（Regime 1 至 3）中检索校准缓冲，预测器与非共形分数保持不变。
正文
Abstract:Conformal prediction guarantees marginal coverage, but a pooled calibration quantile can hide systematic undercoverage across heterogeneous regions of the feature space. We introduce Self-Organized Conformal Prediction (SOCP), a calibration scheme that discovers input-space groups with an unsupervised Self-Organizing Map (SOM) trained without calibration data. At prediction time, the query's best-matching unit (BMU) retrieves a calibration buffer from one cell, a fixed grid neighborhood, or a prototype enlargement (Regimes 1 to 3). When fixed neighborhoods are too sparse, Regime 3 adds cells by prototype distance, using a global budget selected from training cell occupancies and the planned calibration size before any calibration score is observed. The predictor and nonconformity score remain unchanged. Cell-only retrieval has exact cell-conditional validity, and each fixed union of cells has exact retrieved-set validity. Interpreting a neighborhood threshold at its central cell incurs an explicit Kolmogorov-Smirnov (KS) bias term. Across ten regression and classification benchmarks, SOCP lowers the mean weighted coverage gap (WCovGap) relative to pooled split conformal prediction (SCP) on all ten datasets. The mean relative change is -14.8%, with a mean increase of 3.0% in output size. Under fixed-neighborhood retrieval, SO calibration lowers mean WCovGap across ten seeds in 43 of 50 score and dataset comparisons, while SO-SCP lowers it for every dataset at all three tested external partition granularities. These results provide local group calibration without supervised partitions or predictor retraining, while the diagnostics make the cost and limits of locality explicit. The code is available at this https URL.
| Comments: | Accepted at NeurIPS 2026. Updated to the camera-ready version |
| Subjects: | Machine Learning (stat.ML); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2606.29403 [stat.ML] |
| (or arXiv:2606.29403v3 [stat.ML] for this version) | |
| https://doi.org/10.48550/arXiv.2606.29403 arXiv-issued DOI via DataCite |
Submission history
From: Louis Berthier [view email]
[v1]
Sun, 28 Jun 2026 13:50:13 UTC (2,108 KB)
[v2]
Mon, 31 Aug 2026 11:32:23 UTC (838 KB)
[v3]
Tue, 13 Oct 2026 18:02:06 UTC (834 KB)
来源：arXiv：cs.LG · arxiv.org