Fisher Information and Support Complexity of Gaussian Mixtures

Abstract

We study lower bounds on the Fisher information of Gaussian location mixtures and their applications to estimation problems. First, for a discrete mixing distribution, we derive a universal positive-semidefinite lower bound on the Fisher information matrix in terms of the Rényi entropy of order two. For a mixing distribution supported on K points, this bound scales as log(K)/K², improving by a logarithmic factor on the support-size dependence obtained from Stam's inequality. We then develop dimension-aware trace bounds using classical Sobolev inequalities. These bounds are expressed in terms of Sibson's α-mutual information and yield the scaling K^−2/n in dimension n. A complementary bound involving the α-mutual information of order 1/2 is obtained using the sharp Young convolution inequality. The derived inequalities are further specialized to rotationally invariant mixing distributions supported on discrete spherical shells. In this setting, an intrinsic-dimensional bound on Sibson mutual information leads to lower bounds on the Fisher information in terms of the number of occupied shells. As applications, we study least favorable priors for bounded Gaussian mean estimation. In the scalar case, we show that the least favorable prior on [−R,R] has at least Ω(R√(log R)) support points and that each of its probability masses is of order at most 1/R. In the vector case, we show that the number of occupied shells grows at least linearly with R. Finally, Brown's identity and the I-MMSE relationship are used to translate the Fisher-information inequalities into upper bounds on the MMSE and mutual information.

Thermal Quantum Sensing: Fisher Information and Work Beyond Gaussian Signals
Yash Chitgopekar, Nikolaos Koukoulekidis, Iman Marvian
arXiv9 Sep 2026PreprintarXiv licenceFull textMathematical physics Atomic, molecular & optical physicsarXiv
Dimension-free comparison inequalities between f-divergences for Gaussian location mixtures
Jeonghwan Lee, Cong Ma
arXiv18 Oct 2026PreprintarXiv licenceFull textProbability StatisticsarXiv
On the discrete Fisher information
Keith M. Ball, Barnabás Gárgyán
arXiv13 Oct 2026PreprintCC BY 4.0Full textProbabilityarXiv
Solving the Ho--Nguyen Conjecture for Gaussian Mixtures: Applications to Mixtures of Experts
Yuyang Zhang, Qihuang Zhang
arXiv19 Oct 2026PreprintarXiv licenceFull textProbability StatisticsarXiv
Squarefree values of random polynomials are Gaussian mixtures
Omer Simhi
arXiv21 Oct 2026PreprintarXiv licenceFull textAlgebra & number theory ProbabilityarXiv

Cite

BibTeX RIS

BibTeX

@misc{barletta2026fisher,
  title = {{Fisher Information and Support Complexity of Gaussian Mixtures}},
  author = {Luca Barletta and Alex Dytso},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.21726},
  eprint = {2610.21726},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.21726},
}

RIS

TY  - UNPB
TI  - Fisher Information and Support Complexity of Gaussian Mixtures
AU  - Luca Barletta
AU  - Alex Dytso
PY  - 2026
DA  - 2026/10/15
PB  - arXiv
DO  - 10.48550/arxiv.2610.21726
UR  - https://arxiv.org/abs/2610.21726
AB  - We study lower bounds on the Fisher information of Gaussian location mixtures and their applications to estimation problems. First, for a discrete mixing distribution, we derive a universal positive-semidefinite lower bound on the Fisher information matrix in terms of the Rényi entropy of order two. For a mixing distribution supported on K points, this bound scales as log(K)/K^2, improving by a logarithmic factor on the support-size dependence obtained from Stam's inequality. We then develop dimension-aware trace bounds using classical Sobolev inequalities. These bounds are expressed in terms of Sibson's α-mutual information and yield the scaling K^−2/n in dimension n. A complementary bound involving the α-mutual information of order 1/2 is obtained using the sharp Young convolution inequality. The derived inequalities are further specialized to rotationally invariant mixing distributions supported on discrete spherical shells. In this setting, an intrinsic-dimensional bound on Sibson mutual information leads to lower bounds on the Fisher information in terms of the number of occupied shells. As applications, we study least favorable priors for bounded Gaussian mean estimation. In the scalar case, we show that the least favorable prior on [−R,R] has at least Ω(R√(log R)) support points and that each of its probability masses is of order at most 1/R. In the vector case, we show that the number of occupied shells grows at least linearly with R. Finally, Brown's identity and the I-MMSE relationship are used to translate the Fisher-information inequalities into upper bounds on the MMSE and mutual information.
ER  -

Abstract

Related

Cite