← 研踪 PaperTrailarXiv:2609.36208 · v2
Representable but Unlearned: Encoding Rank and the Interaction-Prediction Floor
首次提交 2026-09-29 04:09
版本更新 2026-11-17 07:26
雷达观测 2026-11-18 19:43
摘要
Input encodings can prevent a predictor from reproducing measured interactions jointly, even when each interaction is individually representable. We compute the attainable contrast space from the encoder and contrast design, without labels or predictor fitting. Projecting measured contrasts onto this space gives an empirical squared-error floor for any deterministic decoder using the encoding. On an siRNA panel, a graph neural network's training-only feature mask merges 165 measured states into 90 classes, reducing the rank of 140 interaction contrasts to 72. The floor is 0.009980, or 14.63% of recorded contrast energy, the mean squared error of predicting zero for every contrast. Its 14.6% share of fitted interaction error adds little: predicted interactions are nearly zero, and their MSE of 0.068335 slightly exceeds that of the same control. Antisense-marginal shares are larger: 36.49% of recorded energy and 38.1% of fitted error. Three restored chemistry columns suffice for full rank. Removing the full mask eliminates the floor, but refitting improves interaction MSE by only 0.000017; the restored features remain absent from training inputs. Two further panels clarify the diagnostic's scope. A released splicing predictor encodes injectively and retains the full design rank of 1,986, so its floor is zero. For an independent count-fingerprint encoder on a lipid panel, 800 pairwise-distinct encoded contrast rows have rank 740, exposing 60 restrictions missed by duplicate-row checks. The diagnostic identifies encoding-imposed limits and can trace them to specific features. It separates representability from learned accuracy, but does not by itself predict refitting gains or explain the remaining error.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-17 07:26 · 最新观测版本
Representable but Unlearned: Encoding Rank and the Interaction-Prediction Floor
Input encodings can prevent a predictor from reproducing measured interactions jointly, even when each interaction is individually representable. We compute the attainable contrast space from the encoder and contrast design, without labels or predictor fitting. Projecting measured contrasts onto this space gives an empirical squared-error floor for any deterministic decoder using the encoding. On an siRNA panel, a graph neural network's training-only feature mask merges 165 measured states into 90 classes, reducing the rank of 140 interaction contrasts to 72. The floor is 0.009980, or 14.63% of recorded contrast energy, the mean squared error of predicting zero for every contrast. Its 14.6% share of fitted interaction error adds little: predicted interactions are nearly zero, and their MSE of 0.068335 slightly exceeds that of the same control. Antisense-marginal shares are larger: 36.49% of recorded energy and 38.1% of fitted error. Three restored chemistry columns suffice for full rank. Removing the full mask eliminates the floor, but refitting improves interaction MSE by only 0.000017; the restored features remain absent from training inputs. Two further panels clarify the diagnostic's scope. A released splicing predictor encodes injectively and retains the full design rank of 1,986, so its floor is zero. For an independent count-fingerprint encoder on a lipid panel, 800 pairwise-distinct encoded contrast rows have rank 740, exposing 60 restrictions missed by duplicate-row checks. The diagnostic identifies encoding-imposed limits and can trace them to specific features. It separates representability from learned accuracy, but does not by itself predict refitting gains or explain the remaining error.
查看此版本解读 →查看此版本原文 ↗