diff options
Diffstat (limited to 'RESULTS.md')
| -rw-r--r-- | RESULTS.md | 33 |
1 files changed, 27 insertions, 6 deletions
@@ -635,12 +635,33 @@ The executable post-failure diagnosis finds a specific precursor rather than only a bad endpoint. Before nonfiniteness, the apical prediction--target cosine never exceeds `8.51e-5` in magnitude, while calibration MSE differs from target power by at most `6.96e-8` relatively. Target power grows `2560.7x` from epoch -1 to 80 and `2.13e9x` by epoch 87. Thus the positive short-run teaching -alignment cannot be attributed to successful causal calibration of `A`; the -full-spatial K1 estimator is effectively all noise after projection into the -small channel-gated vectorizer. This is a post-failure diagnosis, not a passed -preregistered result, and motivates a separately labelled representable- -subspace estimator rather than a threshold change to Oral-A v1. +1 to 80 and `2.13e9x` by epoch 87. These instantaneous metrics show that +stochastic unit targets dominate the regression signal, but the later v2 +experiment also establishes an important interpretation limit: near-zero +single-event target cosine does not by itself prove zero conditional learning. + +The separately frozen post-failure v2 screen perturbs only the two basis fields +expressible by the channel-gated vectorizer. At identical query count and 400 +frozen-forward calibration minibatches, selected `eta_A=0.01` gives: + +| estimator | calibration metric space | early-third alignment | all-layer alignment | +|:--|:--|--:|--:| +| unit targets | full hidden field | 0.001105 | 0.011664 | +| channel subspace | channel-basis moments | 0.007209 | 0.052740 | + +Thus representable-subspace perturbation improves early/all-layer +exact-gradient alignment by 6.5x/4.5x at matched causal-query count. +However, it misses the frozen early alignment threshold (`0.007209 < 0.01`) +and the required absolute advantage (`0.006104 < 0.01`). V2-1 therefore fails, +the full v2 ResNet run is not launched, and confirmation remains untouched. +The useful conclusion is narrower: spatial projection was a real variance +problem, but removing it does not yet deliver sufficient early-layer causal +credit at standard depth. + +The two target-power logs are not directly comparable: unit targets are scored +per full hidden field, whereas structured targets are scored per channel-basis +moment. No empirical 1133x variance-reduction claim is made from their raw +ratio. ## How to run `experiments/run.py --mode {bp,fa,dfa,sdil} --dataset {mnist,fmnist,cifar10} --depth D --residual {0,1} --act {tanh,gelu,silu,relu}` |