Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..

Agnostic Learning under Targeted Poisoning: Optimal Rates and the Role of Randomness

Authors: Bogdan Chornomaz, Yonatan Koren, Shay Moran, Tom Waknine

NeurIPS 2025 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical The paper discusses "Optimal Rates and the Role of Randomness", focuses on "the problem of learning in the presence of an adversary", "resolve the corresponding question in the agnostic setting", "prove Theorem 1 in a form of Theorem 4 below, which gives quantitative version of the bounds announced in it." It contains lemmas, theorems, and proofs. The NeurIPS Paper Checklist explicitly states "The paper does not include experiments."
Researcher Affiliation Collaboration Bogdan Chornomaz Department of Mathematics Technion Israel Institute of Technology Haifa, Israel EMAIL. Shay Moran Departments of Mathematics, Computer Science, and Data and Decision Sciences Technion Israel Institute of Technology and Google Research Haifa, Israel EMAIL.
Pseudocode Yes Putting everything together, our final algorithm proceeds as follows: 1. Given a sample S, draw a random sub-sample T of size O( p VC(H)/η). 2. Use T to construct a finite ε-cover HT H, where ε = p VC(H)/η. 3. Run exponential sampling over HT : select h HT with probability proportional to exp( λ ˆLS(h)).
Open Source Code No Justification: The paper does not include experiments requiring code. Justification: The paper does not release new assets
Open Datasets No Justification: The paper does not include experiments.
Dataset Splits No Justification: The paper does not include experiments.
Hardware Specification No Justification: The paper does not include experiments.
Software Dependencies No Justification: The paper does not include experiments.
Experiment Setup No Justification: The paper does not include experiments.