diff options
Diffstat (limited to 'paper')
| -rw-r--r-- | paper/ONE_PAGE.html | 6 | ||||
| -rw-r--r-- | paper/ONE_PAGE.pdf | bin | 70587 -> 71203 bytes |
2 files changed, 5 insertions, 1 deletions
diff --git a/paper/ONE_PAGE.html b/paper/ONE_PAGE.html index 777e38f..d39017e 100644 --- a/paper/ONE_PAGE.html +++ b/paper/ONE_PAGE.html @@ -71,6 +71,10 @@ control reach 10.38% and 10.31% accuracy. SDIL reaches <b>97.35%</b>. Across five seeds, its paired gain over the magnitude-matched control is <b>87.038 ± 0.607 points</b>. Matching signal size does not explain the result; subtraction changes the direction used for learning.</p> + <p>In a new author-code Dual Prop screen, activity-dependent bias makes raw learning nonfinite + in epoch 1 at all three strengths. Innovation remains finite and reaches 71.10%, 70.18%, and + 68.96%, within 0.90 points of exact bias subtraction. The core comparison passes, but the + complete development gate fails its clean/common endpoint controls.</p> </div> <div> <h2>Accuracy versus cost</h2> @@ -83,7 +87,7 @@ <h2>What would make the paper stronger</h2> <ul> - <li>Show the same failure and recovery inside strong contrastive learners; the frozen Dual Prop bias screen is running now.</li> + <li>Repeat the Dual Prop recovery across seeds after freezing a corrected same-path identity control.</li> <li>Finish the matched CNN, ResNet, and Transformer comparison; report accuracy, runtime, memory, and update cost together.</li> <li>Test bias that drifts or is only partly predictable. Random noise is a negative control because this subtraction cannot remove it.</li> </ul> diff --git a/paper/ONE_PAGE.pdf b/paper/ONE_PAGE.pdf Binary files differindex 4e73b8c..6f973ad 100644 --- a/paper/ONE_PAGE.pdf +++ b/paper/ONE_PAGE.pdf |