Generalized Shape Metrics on Neural Representations
Authors: Alex H Williams, Erin Kunz, Simon Kornblith, Scott Linderman
NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate these methods on large-scale datasets from biology (Allen Institute Brain Observatory) and deep learning (NAS-Bench-101). In doing so, we identify relationships between neural representations that are interpretable in terms of anatomical features and model performance. |
| Researcher Affiliation | Collaboration | Alex H. Williams Statistics Department Stanford University ahwillia@stanford.edu Erin Kunz Electrical Engineering Department Stanford University ekunz@stanford.edu Simon Kornblith Google Research, Toronto skornblith@google.com Scott W. Linderman Statistics Department Stanford University scott.linderman@stanford.edu |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide concrete access to source code (specific repository link, explicit code release statement, or code in supplementary materials) for the methodology described in this paper. |
| Open Datasets | Yes | We analyzed two large-scale public datasets spanning neuroscience (Allen Brain Observatory, ABO; Neuropixels visual coding experiment; [21]) and deep learning (NAS-Bench-101; [22]). |
| Dataset Splits | No | The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) needed to reproduce the data partitioning for its own analysis/modeling tasks. |
| Hardware Specification | No | The paper does not provide specific hardware details (exact GPU/CPU models, processor types, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper mentions software like "Scikit-learn" and "Geomstats" but does not provide specific version numbers for these or any other software dependencies. |
| Experiment Setup | Yes | The embedding dimension L is a user-defined hyperparameter. ... Our simple approach already yields promising results: we find that moderate embedding dimensions (L 20) is sufficient to produce high-quality embeddings. |