|
Download lean/docs/075.md from roseking/openAI-math-dataset: direct link, hf CLI and curl.
- Browser
- Download file 964 Bytes
-
https://huggingface.co/datasets/roseking/openAI-math-dataset/resolve/main/lean/docs/075.md
- Command line
-
hf download hf://datasets/roseking/openAI-math-dataset/lean/docs/075.md
-
curl -L -o 075.md https://huggingface.co/datasets/roseking/openAI-math-dataset/resolve/main/lean/docs/075.md
964 Bytes
| # The $`L\log L`$ Fourier-convergence conjecture | |
| The following describes the scope of the Lean formalization related to the following accompanying paper(s): | |
| - [Almost-everywhere Fourier convergence in L log L](../../preprints/Almost-everywhere-Fourier-convergence-in-L-log-L-September-23-2026/paper.pdf) | |
| ## Scope | |
| The $L\log L$ Fourier-convergence conjecture asks whether ordinary symmetric Fourier partial sums converge almost everywhere at this integrability scale. The formalization proves this for every measurable complex-valued function on $\mathbb T=\mathbb R/(2\pi\mathbb Z)$ satisfying $\int_{\mathbb T}|f|\log(2+|f|)<\infty$. Outside one measurable null set, | |
| $\sum_{k=-N}^{N}\widehat f(k)e^{ikx}\to f(x)$ along the full sequence $N\to\infty$. | |
| ## Comparator links | |
| | Result | Comparator statement | | |
| | --- | --- | | |
| | Almost-everywhere convergence of full symmetric Fourier partial sums | [FourierLLogL.lean](../ComparatorChallenges/FourierLLogL.lean) | | |