diff options
| author | YurenHao0426 <Blackhao0426@gmail.com> | 2026-05-28 23:19:05 -0500 |
|---|---|---|
| committer | YurenHao0426 <Blackhao0426@gmail.com> | 2026-05-28 23:19:05 -0500 |
| commit | 66dc2d38ad8187e111843b911d29b9fc23283aa0 (patch) | |
| tree | bbd5ffb3f6499b2c122dab7d6d4b7ca9683effc6 /notes/03_paper_outline.md | |
| parent | 0a3cc65f54d4ffc90b0e73f1f8713820352b27bf (diff) | |
Add trajectory ensemble validation
Diffstat (limited to 'notes/03_paper_outline.md')
| -rw-r--r-- | notes/03_paper_outline.md | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/notes/03_paper_outline.md b/notes/03_paper_outline.md index a8f07c8..65a5c47 100644 --- a/notes/03_paper_outline.md +++ b/notes/03_paper_outline.md @@ -176,6 +176,7 @@ Initial implementation note: - Start with synthetic full-batch regression and no-bias MLPs. - Record same-point gradient alignment by computing BP and FA gradients at the FA weights. - Treat early trajectory runs as instrumentation checks before drawing distributional conclusions. +- Use within-architecture correlations as the first trajectory validation target; pooled correlations can be misleading because architecture-level task difficulty and BP final loss vary. ## 9. Discussion |
