# Minimax Rates for Estimating the Dimension of a Manifold

arXiv: https://arxiv.org/abs/1605.01011 · HTML: https://mlsanity.com/paper/1605.01011/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 May 2016.

## Built on by (most cited first, 10 of 10)

- [The Intrinsic Dimension of Images and Its Impact on Learning](https://mlsanity.com/paper/2104.08894/) (2021)
- [Manifold learning: what, how, and why](https://mlsanity.com/paper/2311.03757/) (2023)
- [Intrinsic Dimension Estimation Using Wasserstein Distances](https://mlsanity.com/paper/2106.04018/) (2021)
- [Tangent Space and Dimension Estimation with the Wasserstein Distance](https://mlsanity.com/paper/2110.06357/) (2021)
- [Nonparametric regression on random geometric graphs sampled from submanifolds](https://mlsanity.com/paper/2405.20909/) (2024)
- [Statistical embedding: Beyond principal components](https://mlsanity.com/paper/2106.01858/) (2021)
- [Reproducibility and Geometric Intrinsic Dimensionality: An Investigation on Graph Neural Network Research](https://mlsanity.com/paper/2403.08438/) (2024)
- [On consistent estimation of dimension values](https://mlsanity.com/paper/2412.13898/) (2024)
- [A Survey and Comparative Evaluation of Intrinsic Dimension Estimators under the Manifold Hypothesis](https://mlsanity.com/paper/2509.15517/) (2025)
- [Minimax Lower Bound for Estimating Diffusion-based Local Intrinsic Dimension](https://mlsanity.com/paper/2609.04822/) (2026)

## Abstract (arXiv)

Many algorithms in machine learning and computational geometry require, as input, the intrinsic dimension of the manifold that supports the probability distribution of the data. This parameter is rarely known and therefore has to be estimated. We characterize the statistical difficulty of this problem by deriving upper and lower bounds on the minimax rate for estimating the dimension. First, we consider the problem of testing the hypothesis that the support of the data-generating probability distribution is a well-behaved manifold of intrinsic dimension d1 versus the alternative that it is of dimension d2, with d1<d2. With an i.i.d. sample of size n, we provide an upper bound on the probability of choosing the wrong dimension of Oleft( n^-left(d2/d1-1-εright)n right), where ε is an arbitrarily small positive number. The proof is based on bounding the length of the traveling salesman path through the data points. We also demonstrate a lower bound of Omega left( n^-(2d2-2d1+ε)n right), by applying Le Cam's lemma with a specific set of d1-dimensional probability distributions. We then extend these results to get minimax rates for estimating the dimension of well-behaved manifolds. We obtain an upper bound of order O left( n^-(frac1m-1-ε)n right) and a lower bound of order Omega left( n-(2+ε)n right), where m is the embedding dimension.