关于「东西方原典考辨:《老子·第七十一章》「知不知」与拉普拉斯《概率的哲学随笔》」,《最大熵需求重建》给出的定量结论是什么?
结论:第七章考辨《老子》第七十一章「知不知」的王弼本与帛书本异文,以及拉普拉斯《概率的哲学随笔》的不充分理由原则,用散度量出违背已知均值的「冒充的知识」,均匀捷径的差额为 2.385 比特。
数据口径:本书推翻这一常识:正态捷径私加了对称形状与标准差,示意参数下均值 80、下限 30 时它把 0.133 的概率放在不可能的区间,均值 40 时升到 0.412;均匀捷径隐含均值在区间中点,直接违背已知均值。 只承认上下限与均值时,杰恩斯 1957 年提出的最大熵原则给出截断指数分布,示意甜品店的第 90 百分位客单为 146.0 元。
对照数据表
下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 2.385
| H(p*) = 7.080 bits | ||||
| assumed q | mean | H(p*)-H(q) | D(q||p*) | gap |
| b(2,8.8) | 80.00 | 0.251 | 0.251 | 0.000 |
| b(1.2,5.28) | 80.00 | 0.059 | 0.059 | 0.000 |
| two-hump | 80.00 | 0.371 | 0.371 | 0.000 |
| uniform[30,130] | 80.00 | 0.436 | 0.436 | 0.000 |
| normal(80,45) cut | 91.17 | -0.054 | 0.259 | 0.313 |
| uniform[30,300] | 165.00 | -0.997 | 1.387 | 2.385 |
查看完整验算程序
# Chapter 7. p72: "knowing that one does not know", measured in bits.
# Model: order values on [30, 300] with a stated mean of 80
# (illustrative). For any assumed density q, the divergence
# D(q || p*) from the maximum-entropy answer p* counts the bits q claims
# beyond the stated facts. When q honours the same mean, D equals the
# entropy deficit H(p*) - H(q) exactly (the Pythagorean identity of the
# exponential family); when q breaks the mean, the two part ways and the
# gap flags a claim that contradicts what was said. Simpson rule, fsum.
import math
from statistics import NormalDist
A, B, M = 30.0, 300.0, 80.0
L = B - A
FR = (M - A) / L
def simpson(f, lo, hi, n=2000):
h = (hi - lo) / n
terms = [f(lo), f(hi)]
terms += [4.0 * f(lo + i * h) for i in range(1, n, 2)]
terms += [2.0 * f(lo + i * h) for i in range(2, n, 2)]
return math.fsum(terms) * h / 3.0
def integ(f, n=2000, hi=B):
# x = A + v * v removes the kink of densities that start at A
g = lambda v: f(A + v * v) * 2.0 * v
return simpson(g, 0.0, math.sqrt(hi - A), n)
def te_mean(lam):
u = lam * L
if abs(u) < 1e-8:
return A + L / 2.0
return A + 1.0 / lam - L / math.expm1(u)
def solve(m):
lo, hi = -2.0, 2.0
for _ in range(80):
mid = (lo + hi) / 2.0
if te_mean(mid) > m:
lo = mid
else:
hi = mid
return (lo + hi) / 2.0
def te_pdf(lam):
return lambda x: lam * math.exp(-lam * (x - A)) / -math.expm1(-lam * L)
def te_q(lam, p):
return A - math.log1p(p * math.expm1(-lam * L)) / lam
def beta_pdf(u, al, be):
if u <= 0.0 or u >= 1.0:
return 0.0
lg = math.lgamma(al + be) - math.lgamma(al) - math.lgamma(be)
return math.exp(lg + (al - 1) * math.log(u) + (be - 1) * math.log1p(-u))
def beta_x(al, be):
return lambda x: beta_pdf((x - A) / L, al, be) / L
F2 = (FR - 0.75 * 0.1) / 0.25
MIX = ((0.75, 4.0, 36.0), (0.25, 8.0, 8.0 * (1 - F2) / F2))
def mix_pdf(x):
return math.fsum(w * beta_x(a, b)(x) for w, a, b in MIX)
def truths():
lam = solve(M)
return [("max-ent", te_pdf(lam)),
("b(2,8.8)", beta_x(2.0, 8.8)),
("b(1.2,5.28)", beta_x(1.2, 5.28)),
("two-hump", mix_pdf)]
def ent(q, hi=B):
g = lambda x: -q(x) * math.log2(q(x)) if q(x) > 0.0 else 0.0
return integ(g, 4000, hi)
def kl(q, p, hi=B):
g = lambda x: q(x) * math.log2(q(x) / p(x)) if q(x) > 0.0 else 0.0
return integ(g, 4000, hi)
if __name__ == "__main__":
lam = solve(M)
ps = te_pdf(lam)
h0 = ent(ps)
nd = NormalDist(M, L / 6.0)
z = nd.cdf(B) - nd.cdf(A)
qs = [("b(2,8.8)", beta_x(2.0, 8.8), B),
("b(1.2,5.28)", beta_x(1.2, 5.28), B),
("two-hump", mix_pdf, B),
("uniform[30,130]", lambda x: 0.01, 130.0),
("normal(80,45) cut", lambda x: nd.pdf(x) / z, B),
("uniform[30,300]", lambda x: 1.0 / L, B)]
print("H(p*) = {:.3f} bits".format(h0))
print(" assumed q mean H(p*)-H(q) D(q||p*) gap")
for name, q, hi in qs:
mean = integ(lambda x: x * q(x), 4000, hi)
dh = h0 - ent(q, hi)
d = kl(q, ps, hi)
gap = d - dh if abs(d - dh) >= 5e-4 else 0.0
print(" {:<17} {:>7.2f} {:>11.3f} {:>8.3f} {:>6.3f}".format(
name, mean, dh, d, gap))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《最大熵原理与缺数据需求重建、仅知均价与上下限时最不偏的客单分布推断母典》第 7 章「东西方原典考辨:《老子·第七十一章》「知不知」与拉普拉斯《概率的哲学随笔》」(版本 1.0.0). 5D Library. DOI 待发布
DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-301 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026301,
author = {Fan, Yafu},
title = {The Maximum-Entropy Principle and Demand Reconstruction from Missing Data: the Least-Biased Order-Value Distribution Given Only a Mean and Bounds},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}