From 32123cb36ae9521f60c9b6f67458b931b6540ef2 Mon Sep 17 00:00:00 2001 From: YurenHao0426 Date: Mon, 23 Mar 2026 19:46:08 -0500 Subject: Add final report, plots, experiment guide, and complete NOTE.md All experiments complete: - Toy LQ: credit bridge matches state bridge (~0.94 costate cosine) - CIFAR-10: credit bridge (29.6%) comparable to DFA (30.0%), both beat state bridge (18.5%) - State bridge confirms core hypothesis: perfect state prediction != useful credit - Terminal gradient matching is essential for credit bridge --- report/toy_training_curves.png | Bin 179278 -> 169558 bytes 1 file changed, 0 insertions(+), 0 deletions(-) (limited to 'report/toy_training_curves.png') diff --git a/report/toy_training_curves.png b/report/toy_training_curves.png index cc3532b..e7fedf9 100644 Binary files a/report/toy_training_curves.png and b/report/toy_training_curves.png differ -- cgit v1.2.3