Skip to content
DraconiticResearch
Search Sources API
203,203 studies31 sourcesUpdated 5 min ago
PreprintCC BY 4.0arXiv

DisCo Algebra: Why Disintegration and Combination Are Not Inverses

Abstract

We present DisCo algebra---a compositional framework for joint probability built from a symmetric cooperad (disintegration) and a magma-operad (non-associative partial combination, restricted to the image of disintegration) on the bare family 𝒫n=Prob([0,1]n). No probability integral transform, no external summary functions, no enrichment of objects with conditional kernels. The disintegration side uses block-wise joint CDFs as intrinsic, measure-determined 1D summaries, yielding a lossy compression. The combination side reads the same summaries and reconstructs via regular conditional measures, yielding a lossless embedding on admissible inputs. We prove that combination followed by disintegration is the identity on admissible inputs, but disintegration followed by combination is not---it defines a renormalization operator Rℐ=Combine∘Disintegrate. The information loss of this operator is exactly the probability curvature 𝒦ℐ(μ)---the conditional KL divergence between the true conditional distribution and the conditionally independent product. This curvature measures the obstruction to lossless decomposition and reveals a hard limit on how much cross-block dependence can be captured by CDF-based scalar block summaries.

Related

  1. Complex slices, inverses, and deformations of a three-dimensional real algebra
    Alejandro Jesús Bermejo Valdés
    arXivPreprintCC BY 4.0Full textAlgebra & number theoryarXiv
  2. Palm Disintegration and Height-Profile Recovery for Projected Hardy--Szegő Zeros
    Feng Guo
    arXivPreprintarXiv licenceFull textProbabilityarXiv
  3. On the disintegration of the stochastic majority vote: from pac-bayesian bounds to a self-bounding algorithm
    Julien Bastian, Benjamin Leblanc, Pascal Germain +4
    HALPreprintCC BY-SA 4.0Full textStatisticsHALarXiv
  4. Characterizations in terms of Generalized Inverses
    Feroz Siddique
    arXivPreprintCC BY 4.0Full textAlgebra & number theoryarXiv
  5. Lempert left inverses of the Royal and Flat geodesics of the symmetrized bidisc
    Naveen Gupta
    arXivPreprintCC BY 4.0Full textAnalysis & PDEsarXiv
BibTeX
@misc{lu2026disco,
  title = {{DisCo Algebra: Why Disintegration and Combination Are Not Inverses}},
  author = {Xuexing Lu},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.15006},
  eprint = {2610.15006},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.15006},
  note = {CC BY 4.0},
}
RIS
TY  - UNPB
TI  - DisCo Algebra: Why Disintegration and Combination Are Not Inverses
AU  - Xuexing Lu
PY  - 2026
DA  - 2026/10/21
PB  - arXiv
DO  - 10.48550/arxiv.2610.15006
UR  - https://arxiv.org/abs/2610.15006
AB  - We present DisCo algebra---a compositional framework for joint probability built from a symmetric cooperad (disintegration) and a magma-operad (non-associative partial combination, restricted to the image of disintegration) on the bare family 𝒫_n=Prob([0,1]^n). No probability integral transform, no external summary functions, no enrichment of objects with conditional kernels. The disintegration side uses block-wise joint CDFs as intrinsic, measure-determined 1D summaries, yielding a lossy compression. The combination side reads the same summaries and reconstructs via regular conditional measures, yielding a lossless embedding on admissible inputs. We prove that combination followed by disintegration is the identity on admissible inputs, but disintegration followed by combination is not---it defines a renormalization operator R_ℐ=Combine∘Disintegrate. The information loss of this operator is exactly the probability curvature 𝒦_ℐ(μ)---the conditional KL divergence between the true conditional distribution and the conditionally independent product. This curvature measures the obstruction to lossless decomposition and reveals a hard limit on how much cross-block dependence can be captured by CDF-based scalar block summaries.
ER  -