Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > stat > arXiv:2310.11431

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Statistics > Machine Learning

arXiv:2310.11431 (stat)
[Submitted on 17 Oct 2023 (v1), last revised 18 Oct 2023 (this version, v2)]

Title:Identifying Interpretable Visual Features in Artificial and Biological Neural Systems

Authors:David Klindt, Sophia Sanborn, Francisco Acosta, Frédéric Poitevin, Nina Miolane
View a PDF of the paper titled Identifying Interpretable Visual Features in Artificial and Biological Neural Systems, by David Klindt and 4 other authors
View PDF
Abstract:Single neurons in neural networks are often interpretable in that they represent individual, intuitively meaningful features. However, many neurons exhibit $\textit{mixed selectivity}$, i.e., they represent multiple unrelated features. A recent hypothesis proposes that features in deep networks may be represented in $\textit{superposition}$, i.e., on non-orthogonal axes by multiple neurons, since the number of possible interpretable features in natural data is generally larger than the number of neurons in a given network. Accordingly, we should be able to find meaningful directions in activation space that are not aligned with individual neurons. Here, we propose (1) an automated method for quantifying visual interpretability that is validated against a large database of human psychophysics judgments of neuron interpretability, and (2) an approach for finding meaningful directions in network activation space. We leverage these methods to discover directions in convolutional neural networks that are more intuitively meaningful than individual neurons, as we confirm and investigate in a series of analyses. Moreover, we apply the same method to three recent datasets of visual neural responses in the brain and find that our conclusions largely transfer to real neural data, suggesting that superposition might be deployed by the brain. This also provides a link with disentanglement and raises fundamental questions about robust, efficient and factorized representations in both artificial and biological neural systems.
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG)
Cite as: arXiv:2310.11431 [stat.ML]
  (or arXiv:2310.11431v2 [stat.ML] for this version)
  https://doi.org/10.48550/arXiv.2310.11431
arXiv-issued DOI via DataCite

Submission history

From: David Klindt [view email]
[v1] Tue, 17 Oct 2023 17:41:28 UTC (15,253 KB)
[v2] Wed, 18 Oct 2023 02:02:33 UTC (15,531 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Identifying Interpretable Visual Features in Artificial and Biological Neural Systems, by David Klindt and 4 other authors
  • View PDF
  • TeX Source
  • Other Formats
license icon view license
Current browse context:
stat.ML
< prev   |   next >
new | recent | 2023-10
Change to browse by:
cs
cs.LG
stat

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar
export BibTeX citation Loading...

BibTeX formatted citation

×
Data provided by:

Bookmark

BibSonomy logo Reddit logo

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)

Demos

Replicate (What is Replicate?)
Hugging Face Spaces (What is Spaces?)
TXYZ.AI (What is TXYZ.AI?)

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
  • Author
  • Venue
  • Institution
  • Topic

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?)
  • About
  • Help
  • contact arXivClick here to contact arXiv Contact
  • subscribe to arXiv mailingsClick here to subscribe Subscribe
  • Copyright
  • Privacy Policy
  • Web Accessibility Assistance
  • arXiv Operational Status
    Get status notifications via email or slack