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..
On Agnostic PAC Learning in the Small Error Regime
Authors: Julian Asilis, Mikael Møller Høgsgaard, Grigoris Velegkas
NeurIPS 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | Question: Does the paper discuss the limitations of the work performed by the authors? Answer: [Yes] Justification: We discuss how there is still a small gap in the known lower bounds and upper bounds after our new result. Question: For each theoretical result, does the paper provide the full set of assumptions and a complete (and correct) proof? Answer: [Yes] Justification: Each statement in the paper is proven. Question: Does the paper fully disclose all the information needed to reproduce the main experimental results of the paper to the extent that it affects the main claims and/or conclusions of the paper (regardless of whether the code and data are provided or not)? Answer: [NA] Justification: No experiments are included in the paper. |
| Researcher Affiliation | Collaboration | Julian Asilis USC EMAIL Mikael Møller Høgsgaard Aarhus University EMAIL Grigoris Velegkas Google Research EMAIL |
| Pseudocode | Yes | Algorithm 1: Splitting algorithm S Input: Training sequences S, T (X Y) , where |S| = 3k for k N. Output: Family of training sequences. if k 6 then Partition S into S1, S2, S3, with Si being the (i 1)|S|/3 + 1 to the i|S|/3 training examples of S. Set for each i halloooooooooooooo Si, = Si[1 : 3k 4], Si, = Si[3k 4 + 1 : 3k 1], return [S (S1, ; S1, T), S (S2, ; S2, T), S (S3, ; S3, T)] else Algorithm 2: Splitting algorithm S Input: Training sequences S, T (X Y) , where |S| = 3k for k N. Output: Family of training sequences. if k 6 then Partition S into S1, . . . , S27, with Si being the (i 1)|S|/27 + 1 to the i|S|/27 training examples of S. Set for each i halloooooooooooooo Si, = Si[1 : 3k 6], Si, = Si[3k 6 + 1 : 3k 3], return [S(S1, ; S1, T), . . . , S(S27, ; S27, T)] else |
| Open Source Code | No | Question: Does the paper provide open access to the data and code, with sufficient instructions to faithfully reproduce the main experimental results, as described in supplemental material? Answer: [NA] Justification: No experiments are included in the paper. Question: Are new assets introduced in the paper well documented and is the documentation provided alongside the assets? Answer: [NA] Justification: We do not release any code or dataset. |
| Open Datasets | No | Question: Does the paper provide open access to the data and code, with sufficient instructions to faithfully reproduce the main experimental results, as described in supplemental material? Answer: [NA] Justification: No experiments are included in the paper. Question: Are new assets introduced in the paper well documented and is the documentation provided alongside the assets? Answer: [NA] Justification: We do not release any code or dataset. |
| Dataset Splits | No | No experiments are conducted in this theoretical paper, so there is no mention of dataset splits. |
| Hardware Specification | No | No experiments are conducted in this theoretical paper, so there is no mention of hardware specifications. |
| Software Dependencies | No | No experiments are conducted in this theoretical paper, so there is no mention of software dependencies. |
| Experiment Setup | No | No experiments are conducted in this theoretical paper, so there is no mention of experimental setup details such as hyperparameters or training settings. |