|
Notes on the Problem (Phase 1)
|
|
0
|
3
|
August 19, 2026
|
|
A technique census of Phase 1: what the mathematics is doing, what walls it hit, and what's still open
|
|
12
|
353
|
August 18, 2026
|
|
Phase 1 Substack Post
|
|
0
|
22
|
August 18, 2026
|
|
Phase 2 Starting?
|
|
1
|
30
|
August 18, 2026
|
|
Our current approach: Korobov hybrid trunk (E) + offline ridge residual corrector (G) — ~16% graded gain, what we learned
|
|
0
|
22
|
August 18, 2026
|
|
Phase 1 Report, Submission 327801
|
|
1
|
36
|
August 18, 2026
|
|
Measured: the metered-FLOP vs residual-wall-time trade-off, and what it implies for a Phase 2 cap
|
|
0
|
22
|
August 18, 2026
|
|
Bitter-lesson-pilled sampling-free propagator yielding 1.5e-5 raw MSE
|
|
3
|
98
|
August 17, 2026
|
|
[Write-up, submission 326725] Measuring the ceiling of a method class: how we closed 54 families of estimators without building them
|
|
0
|
60
|
August 17, 2026
|
|
Phase 1 Algorithmic Contribution Write-up — Submission #327792
|
|
0
|
46
|
August 17, 2026
|
|
Phase 1 Write-Ups, Team Puffi
|
|
3
|
102
|
August 17, 2026
|
|
Phase 1 Write-Up: Useful tools, rejected ideas, a warning, and more
|
|
0
|
58
|
August 17, 2026
|
|
[Phase 1 write-up] A calibrated account of the sampling plateau, plus two negative results (submission #326954)
|
|
0
|
66
|
August 17, 2026
|
|
Phase 1 write-up: The missing basis was not the mechanism - submission #327749
|
|
0
|
74
|
August 17, 2026
|
|
Phase I Submission: Structure-Aware Estimation for Random ReLU MLPs
|
|
0
|
971
|
July 27, 2026
|
|
[Phase 1 Write-Up] My Write-Up for Submission 326960
|
|
0
|
64
|
August 16, 2026
|
|
[Phase 1 Write-Up] My Approach Write-Up for the submission 327651
|
|
0
|
93
|
August 16, 2026
|
|
[Phase 1 writeup] Fixed positive sampling is nearly saturated (submission 327786)
|
|
0
|
101
|
August 16, 2026
|
|
Phase 1 Write Up - Bias–Variance Tradeoff Across Control Layers (submission #327723)
|
|
0
|
97
|
August 15, 2026
|
|
Phase 1 skibidi toilet solution writeup
|
|
2
|
170
|
August 15, 2026
|
|
Phase 1 Ride and Personal Insights on my trials
|
|
0
|
86
|
August 15, 2026
|
|
Current top 9: Notes on Phase 1 board-score variance, the truth-energy ladder, and four corrections to published figures
|
|
0
|
225
|
August 15, 2026
|
|
[Visual notebook] How 203 experiments reached public #16: the complete idea map
|
|
0
|
108
|
August 14, 2026
|
|
[Phase 1 research atlas] How I reached public #16 — and why the next 56 experiments failed
|
|
2
|
171
|
August 14, 2026
|
|
[Phase 1 write-up] Public #16 → #40 despite a better score: 203 investigations, one lawful survivor (#326680)
|
|
0
|
389
|
August 11, 2026
|
|
Phase 1 Write-Up: A Modern Approach
|
|
1
|
269
|
August 12, 2026
|
|
Phase 1 write-up: Monetizing the approach to the moment-propagation frontier — an intra-network control variate, and a measured map of the wall behind it (submission #326024)
|
|
0
|
271
|
August 12, 2026
|
|
Phase 1 post-mortem
|
|
1
|
298
|
August 11, 2026
|
|
Phase 1 write-up: Stabilizing cumulant propagation at depth 32 — a trajectory-calibrated moment chain, with an error budget (submission #314695)
|
|
3
|
1389
|
August 11, 2026
|
|
[Phase 1 write-up] Where the closure family's floor actually is, and what the residual term costs per call (submission 324409)
|
|
0
|
342
|
August 11, 2026
|