SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2606.00342 · 2026

PE-means: Improved Differentially Private 𝑘-means Clustering through Private Evolution

Sergey Yekhanin, Zinan Lin, Thomas Humphries

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 9 functions out of this paper's own repositories and ran 9 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
t3humphries/PE-means canonical 9 of 9
FunctionStatusWhere it lives
compute_epsilon Ran t3humphries/PE-means/src/DP_comp.py
code served (permissive licence) · get_code("e2d6f607fcde45e2")
dataset_preprocess Ran t3humphries/PE-means/src/helpers.py
code served (permissive licence) · get_code("d9155bf332092151")
delta_Gaussian Ran t3humphries/PE-means/src/DP_comp.py
code served (permissive licence) · get_code("32fa80153875de17")
eps_Gaussian Ran t3humphries/PE-means/src/DP_comp.py
code served (permissive licence) · get_code("773e62a14de7104f")
get_labels_memory_efficient Ran t3humphries/PE-means/src/pe_means_main.py
code served (permissive licence) · get_code("fa200e45ebd8daf9")
levy_mutation Ran t3humphries/PE-means/src/pe_means_main.py
code served (permissive licence) · get_code("27d2fedf6b91f95c")
load_txt Ran t3humphries/PE-means/src/helpers.py
code served (permissive licence) · get_code("2a9d75d1feca994c")
run_single_iteration Ran t3humphries/PE-means/baselines/icml_baseline.py
code served (permissive licence) · get_code("76a4d233387691ad")
simple_loss Ran t3humphries/PE-means/src/helpers.py
code served (permissive licence) · get_code("9ffd334fedc7e7c3")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We study the problem of differentially private (DP) 𝑘-means clustering in Euclidean space. Previous solutions rely on summing the private data directly, which induces a sensitivity proportional to the domain. We introduce PE-means, an extension of the private evolution (PE) algorithm (an increasingly popular method for synthetic data generation), to the problem of 𝑘-means clustering. The key advantage of PE is that it only computes a private histogram with constant sensitivity to guide the evolution. Our adaptation of PE includes new evolutionary operators for clustering, as well as other algorithmic improvements of independent interest. Overall, PE-means achieves an average improvement of 26% in clustering loss over state-of-the-art baselines such as Google's LSH-based algorithm and DP-Lloyd variants.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2606.00342")
get_code_for_paper("2606.00342")
have("2606.00342")

Connect an agent — have() is free.