Skip to main content

Follow the source behind the result.

Search the paper sections, open their original context and follow the checks they actually cite. The knowledge atlas explains the result; this library locates its source.

v23_anchor_generator.py

verification/v23_anchor_generator.py

This is a registered verification module. Read its original code and the paper sections that cite it, then follow the linked claim to its present classification.

Read the publication code on this page

The registered Python source is shown as text. Displaying it does not execute the code.

"""v23 -- Anchor-first generator: a=(1,1,2) generates the two axioms and E8.

Reviewer insight (TFPT 5.1): the parabolic flavor anchor a=(1,1,2) -- the same
(1,1,2) that is the exponents-at-infinity / splitting type O(-2)+O(-1)^2 -- is
itself the generator of the whole compiler.  Its elementary symmetric
polynomials give the two axioms, and its power sums generate the big Lie data:

  elementary symmetric e_k(a) = (4, 5, 2) = (|mu4|, g_car, |Z2|)
      => c3 = 1/(2 e_1 pi) = 1/(8 pi)   and   g_car = e_2 = 5
  power sums p_n(a) = 1^n + 1^n + 2^n = 2 + 2^n:
      p_1=4=|mu4|, p_2=6=|R^+(A3)|, p_3=10=A_Lambda,
      |R(E8)| = p_1 p_2 p_3 = 240,
      dim E8  = p_1 p_2 p_3 + (p_4 - p_3) = 248,
      p_4 - p_3 = 8 = rank E8,
      p_{n+1} - p_n = 2^n (the binary ladder).

So the inputs reduce from {c3, g_car} to {a=(1,1,2), pi}.
"""
import sympy as sp
from tfpt_constants import check, summary, reset, g_car


def esym(a):
    x = sorted(a)
    e1 = sum(x)
    e2 = x[0]*x[1] + x[0]*x[2] + x[1]*x[2]
    e3 = x[0]*x[1]*x[2]
    return e1, e2, e3


def psum(a, n):
    return sum(v**n for v in a)


def run():
    reset()
    print("v23  anchor-first generator a=(1,1,2)")
    a = (1, 1, 2)

    # elementary symmetric polynomials -> the two axioms
    e1, e2, e3 = esym(a)
    check("e_1(a) = 4 = |mu4|", e1, 4, exact=True)
    check("e_2(a) = 5 = g_car", e2 == g_car)
    check("e_3(a) = 2 = |Z2| (sheet)", e3, 2, exact=True)
    check("c3 = 1/(2 e_1 pi) = 1/(8 pi)", sp.Rational(1, 2 * e1) / sp.pi == sp.Rational(1, 8) / sp.pi)

    # power sums p_n = 2 + 2^n -> the big Lie data
    p = {n: psum(a, n) for n in range(1, 6)}
    check("p_n(a) = 2 + 2^n", [p[n] for n in range(1, 6)], [2 + 2**n for n in range(1, 6)], exact=True)
    check("p_1 = 4 = |mu4|", p[1], 4, exact=True)
    check("p_2 = 6 = |R^+(A3)|", p[2], 6, exact=True)
    check("p_3 = 10 = A_Lambda", p[3], 10, exact=True)
    check("|R(E8)| = p_1 p_2 p_3 = 240", p[1] * p[2] * p[3], 240, exact=True)
    check("dim E8 = p_1 p_2 p_3 + (p_4 - p_3) = 248", p[1] * p[2] * p[3] + (p[4] - p[3]), 248, exact=True)
    check("p_4 - p_3 = 8 = rank E8", p[4] - p[3], 8, exact=True)
    check("binary ladder p_{n+1} - p_n = 2^n", [p[n + 1] - p[n] for n in range(1, 5)],
          [2**n for n in range(1, 5)], exact=True)

    # the anchor IS the flavor anchor (exponents at infinity / splitting type)
    check("a=(1,1,2) = exponents at infinity of O(-2)+O(-1)^2 (the flavor anchor)",
          sorted(a) == [1, 1, 2])
    return summary("v23 anchor generator")


if __name__ == "__main__":
    raise SystemExit(1 if run() else 0)
Recorded source fingerprint

e782df61e9190f7087d3c93e2524388a037784f7d21d2a251399f195357982dd

Paper sections that cite this module

The module’s catalogue entry retains all 13 section references.

What this source index covers

The public registry includes 116 document views, 6,693 section records and 985 physical LaTeX sources. Active publication documents, research material and archives retain their recorded roles. This is source navigation, not a new semantic review or proof run.