Skip to content

[Grok retrieval] Double-pulse / bilinear kernels vs P398 odd-sector certificate #713

Description

@LightChainr

Parent: #650. Retrieval-only. No new P398 production.

Why

Draft #709: on the existing P398 join/detach calibration (not percolation), two odd insertions make the delay kernel S H exp(τG) H F exact order 4/16 at widths 4/5, exhausting the odd subspace, while one insertion is identically zero by parity. Width-4 wrapped-pair/wrap scalar Hankel det = -16.

Independent literature: bilinear/Volterra kernels, Schur-complement memory, and partition-process generators.

Deliverable

notes/lit-double-pulse-observability-YYYYMMDD.md as a draft PR against main. No STATUS. Do not merge.

Questions

  1. Markov chains / interacting particle systems: two-pulse or Volterra kernels used to recover a hidden odd sector after an even quotient. Primary papers, not textbooks only.
  2. Realization theory: bilinear systems, Fliess kernels, Hankel rank of C A^k B for K(τ)=C exp(τA) B. Standard references with theorem numbers. What is not claimed as new if we only certify a finite instance.
  3. Noncrossing partitions / Temperley–Lieb / join-split generators as continuous-time Markov processes (Pitman, Diaconis, or coalescent literature). Is P398's G=sum (J_j+D_j) a named process?
  4. Any published "zero delay misses a delay-mode" example of the same shape as K(0)=0 but Hankel rank 4.

PRIMARY_TEXT_READ / ABSTRACT_ONLY / [LIT]. Tripwire: bilinear/Hankel prior art cited as PRIMARY, plus a yes/no on whether the P398 double-pulse certificate itself appears in print (expected: no).

Wait for ASSIGNED_MACHINE.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    priority:P1Bounded parallel analysis or a concrete reserve direction; not all run at once.

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions