Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
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.
Learning single index models via harmonic decomposition
Authors: Nirmit Joshi, Hugo Koubbi, Theodor Misiakiewicz, Nati Srebro
NeurIPS 2025 | Venue PDF | LLM Run Details | Input Tokens: 68,115 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 2,730 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | In this paper, we characterize the sample and computational complexity of learning single-index models under general spherically symmetric input distributions. ... We establish the following lower bounds on the sample complexity m and runtime T for recovering w using the low-degree polynomial (LDP) and statistical query (SQ) frameworks. ... There are no experimental results, as this is a theory paper. |
| Researcher Affiliation | Academia | Nirmit Joshi Toyota Technological Institute at Chicago EMAIL Hugo Koubbi CEREMADE, UMR 7534, Université Paris Dauphine PSL, Paris , France EMAIL Theodor Misiakiewicz Department of Statistics and Data Science Yale University EMAIL Nathan Srebro Toyota Technological Institute at Chicago EMAIL |
| Pseudocode | Yes | Algorithm 1: A spectral algorithm on the frequency ℓ= 1 and ℓ= 2. ... Algorithm 2: A single step of the boosting algorithm on ℓ 3. ... Algorithm 3: boost-step ... Algorithm 4: Online SGD algorithm on the frequency ℓ. |
| Open Source Code | No | Justification: There are no experimental results, as this is a theory paper. |
| Open Datasets | No | Justification: There are no experimental results, as this is a theory paper. |
| Dataset Splits | No | Justification: There are no experimental results, as this is a theory paper. |
| Hardware Specification | No | Justification: There are no experimental results, as this is a theory paper. |
| Software Dependencies | No | Justification: There are no experimental results, as this is a theory paper. |
| Experiment Setup | No | Justification: There are no experimental results, as this is a theory paper. |