Skip to content
AI News HubLIVE
Original source2 min read

Median-of-Means as an Extremal Convex Estimator and a Nonconvex Route to the Trimmed Oracle

Summary

This paper revisits median-of-means estimation from a deterministic optimization viewpoint, proving that any convex block M-estimator has worst-case robustness constant at least 1/(1-2ε) and thus cannot attain the trimmed-block oracle constant 1/(1-ε); it then introduces a nonconvex block-Lp family whose finite-sample bounds continuously approach the oracle constant as p→0, with global minimizers matching the oracle under mild separation and a benign optimization landscape.

SourcearXiv Machine LearningAuthor: Angshul Majumdar
Median-of-Means as an Extremal Convex Estimator and a Nonconvex Route to the Trimmed Oracle
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 1 Sep 2026]

Title:Median-of-Means as an Extremal Convex Estimator and a Nonconvex Route to the Trimmed Oracle

View a PDF of the paper titled Median-of-Means as an Extremal Convex Estimator and a Nonconvex Route to the Trimmed Oracle, by Angshul Majumdar

View PDF HTML (experimental)

Abstract:We revisit median-of-means estimation from a deterministic optimization viewpoint and develop a family of block-Lp estimators for robust learning with heavy-tailed and adversarially corrupted data. In a block contamination model with at least a fraction 1 minus epsilon of good blocks, we first show that every convex block M-estimator has worst-case robustness constant at least 1 divided by 1 minus 2 epsilon. This matches the classical median-of-means bound and proves that the trimmed-block oracle constant 1 divided by 1 minus epsilon cannot be attained within the convex class. We then introduce a nonconvex block-Lp family for p between 0 and 1 and derive finite-sample deterministic robustness bounds for all global minimizers. As p decreases from 1 toward 0, these bounds continuously approach the trimmed-block oracle constant. For sufficiently small p, the global minimizers coincide with those of the oracle under a mild separation condition. We also show that the block-Lp objectives have a benign landscape, with all local minima remaining close to the truth and no bad basins. Combining these results with block-level concentration yields sub-Gaussian deviation bounds under finite 2 plus delta moments and high-dimensional extensions to robust mean estimation and sparse regression.

Subjects:

Machine Learning (cs.LG)

Cite as: arXiv:2609.01689 [cs.LG]

(or arXiv:2609.01689v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2609.01689

arXiv-issued DOI via DataCite

Journal reference: Majumdar, A. Median-of-Means as an Extremal Convex Estimator and a Nonconvex Route to the Trimmed Oracle. Mach Learn 115, 172 (2026)

Related DOI:

https://doi.org/10.1007/s10994-026-07101-2

DOI(s) linking to related resources

Submission history

From: Angshul Majumdar Dr. [view email] [v1] Tue, 1 Sep 2026 15:39:04 UTC (33 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Median-of-Means as an Extremal Convex Estimator and a Nonconvex Route to the Trimmed Oracle, by Angshul Majumdar

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.LG

new | recent | 2026-09

Change to browse by:

cs

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

IArxiv recommender toggle

IArxiv Recommender (What is IArxiv?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

ResearchersAdvanced

Key points

  • Proves a lower bound of 1/(1-2ε) on the worst-case robustness constant of every convex block M-estimator, matching the classical median-of-means bound and showing the trimmed-block oracle constant 1/(1-ε) is unattainable within the convex class.
  • Introduces a nonconvex block-Lp estimator family for p between 0 and 1, with deterministic finite-sample robustness bounds for all global minimizers that continuously approach the trimmed-block oracle constant as p decreases.
  • Shows the block-Lp objectives have a benign landscape, with all local minima close to the truth and no bad basins; for sufficiently small p, global minimizers coincide with the oracle under a mild separation condition.
  • Combines the results with block-level concentration to obtain sub-Gaussian deviation bounds under finite 2+δ moments, with high-dimensional extensions to robust mean estimation and sparse regression.

Highlights and analysis are generated automatically and may contain errors. Check the original source.