From be39c2b5ebec37f993b1a862459455a98cf39eb2 Mon Sep 17 00:00:00 2001 From: YurenHao0426 Date: Wed, 8 Apr 2026 05:57:53 -0500 Subject: Round 35: SB and CB also show data-agnostic Mode 1 growth on random targets MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - experiments/cifar_resmlp.py: add --methods filter and --random_targets flag; extend compute_diagnostics to log hidden_norms_per_layer and bp_grad_norms_per_layer - paper/main.tex §3 ¶1: broaden random-target finding to all 3 fixed-feedback methods (DFA: ||h_L||=14510, SB: ||h_L||=6225, CB: ||h_L||=19974 at ep 3, all at chance acc) - paper/main.tex Appendix J: extended with cross-method smoke-test table This generalizes the §3 mechanism story from 'DFA-specific' to 'all 3 audited fixed-feedback local-credit methods'. Combined with rounds 32-34, the proximate cause of Mode 1 (a) is now well-localized: - Not requires residual skip (round 33 H2 walkback) - Not requires task signal (round 34 random targets, DFA) - Not DFA-specific (round 35 random targets, SB+CB) Co-Authored-By: Claude Opus 4.6 (1M context) --- paper/main.pdf | Bin 461963 -> 464384 bytes 1 file changed, 0 insertions(+), 0 deletions(-) (limited to 'paper/main.pdf') diff --git a/paper/main.pdf b/paper/main.pdf index 0ba5ae7..a62416c 100644 Binary files a/paper/main.pdf and b/paper/main.pdf differ -- cgit v1.2.3