Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287019225> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4287019225 abstract "Existing research on making sense of deep neural networks often focuses on neuron-level interpretation, which may not adequately capture the bigger picture of how concepts are collectively encoded by multiple neurons. We present NeuroCartography, an interactive system that scalably summarizes and visualizes concepts learned by neural networks. It automatically discovers and groups neurons that detect the same concepts, and describes how such neuron groups interact to form higher-level concepts and the subsequent predictions. NeuroCartography introduces two scalable summarization techniques: (1) neuron clustering groups neurons based on the semantic similarity of the concepts detected by neurons (e.g., neurons detecting dog faces of different breeds are grouped); and (2) neuron embedding encodes the associations between related concepts based on how often they co-occur (e.g., neurons detecting dog face and dog tail are placed closer in the embedding space). Key to our scalable techniques is the ability to efficiently compute all neuron pairs' relationships, in time linear to the number of neurons instead of quadratic time. NeuroCartography scales to large data, such as the ImageNet dataset with 1.2M images. The system's tightly coordinated views integrate the scalable techniques to visualize the concepts and their relationships, projecting the concept associations to a 2D space in Neuron Projection View, and summarizing neuron clusters and their relationships in Graph View. Through a large-scale human evaluation, we demonstrate that our technique discovers neuron groups that represent coherent, human-meaningful concepts. And through usage scenarios, we describe how our approaches enable interesting and surprising discoveries, such as concept cascades of related and isolated concepts. The NeuroCartography visualization runs in modern browsers and is open-sourced." @default.
- W4287019225 created "2022-07-25" @default.
- W4287019225 creator A5020153026 @default.
- W4287019225 creator A5020702008 @default.
- W4287019225 creator A5034842879 @default.
- W4287019225 creator A5036502813 @default.
- W4287019225 creator A5044691513 @default.
- W4287019225 creator A5057757200 @default.
- W4287019225 creator A5080665667 @default.
- W4287019225 date "2021-08-29" @default.
- W4287019225 modified "2023-09-28" @default.
- W4287019225 title "NeuroCartography: Scalable Automatic Visual Summarization of Concepts in Deep Neural Networks" @default.
- W4287019225 doi "https://doi.org/10.48550/arxiv.2108.12931" @default.
- W4287019225 hasPublicationYear "2021" @default.
- W4287019225 type Work @default.
- W4287019225 citedByCount "0" @default.
- W4287019225 crossrefType "posted-content" @default.
- W4287019225 hasAuthorship W4287019225A5020153026 @default.
- W4287019225 hasAuthorship W4287019225A5020702008 @default.
- W4287019225 hasAuthorship W4287019225A5034842879 @default.
- W4287019225 hasAuthorship W4287019225A5036502813 @default.
- W4287019225 hasAuthorship W4287019225A5044691513 @default.
- W4287019225 hasAuthorship W4287019225A5057757200 @default.
- W4287019225 hasAuthorship W4287019225A5080665667 @default.
- W4287019225 hasBestOaLocation W42870192251 @default.
- W4287019225 hasConcept C103278499 @default.
- W4287019225 hasConcept C115961682 @default.
- W4287019225 hasConcept C153180895 @default.
- W4287019225 hasConcept C154945302 @default.
- W4287019225 hasConcept C169760540 @default.
- W4287019225 hasConcept C170858558 @default.
- W4287019225 hasConcept C2778794669 @default.
- W4287019225 hasConcept C36464697 @default.
- W4287019225 hasConcept C41008148 @default.
- W4287019225 hasConcept C41608201 @default.
- W4287019225 hasConcept C48044578 @default.
- W4287019225 hasConcept C50644808 @default.
- W4287019225 hasConcept C73555534 @default.
- W4287019225 hasConcept C77088390 @default.
- W4287019225 hasConcept C80444323 @default.
- W4287019225 hasConcept C86803240 @default.
- W4287019225 hasConceptScore W4287019225C103278499 @default.
- W4287019225 hasConceptScore W4287019225C115961682 @default.
- W4287019225 hasConceptScore W4287019225C153180895 @default.
- W4287019225 hasConceptScore W4287019225C154945302 @default.
- W4287019225 hasConceptScore W4287019225C169760540 @default.
- W4287019225 hasConceptScore W4287019225C170858558 @default.
- W4287019225 hasConceptScore W4287019225C2778794669 @default.
- W4287019225 hasConceptScore W4287019225C36464697 @default.
- W4287019225 hasConceptScore W4287019225C41008148 @default.
- W4287019225 hasConceptScore W4287019225C41608201 @default.
- W4287019225 hasConceptScore W4287019225C48044578 @default.
- W4287019225 hasConceptScore W4287019225C50644808 @default.
- W4287019225 hasConceptScore W4287019225C73555534 @default.
- W4287019225 hasConceptScore W4287019225C77088390 @default.
- W4287019225 hasConceptScore W4287019225C80444323 @default.
- W4287019225 hasConceptScore W4287019225C86803240 @default.
- W4287019225 hasLocation W42870192251 @default.
- W4287019225 hasOpenAccess W4287019225 @default.
- W4287019225 hasPrimaryLocation W42870192251 @default.
- W4287019225 hasRelatedWork W2156856059 @default.
- W4287019225 hasRelatedWork W2244306971 @default.
- W4287019225 hasRelatedWork W2900794075 @default.
- W4287019225 hasRelatedWork W3183633970 @default.
- W4287019225 hasRelatedWork W3202077457 @default.
- W4287019225 hasRelatedWork W3207688490 @default.
- W4287019225 hasRelatedWork W3208309985 @default.
- W4287019225 hasRelatedWork W4207080266 @default.
- W4287019225 hasRelatedWork W4214772220 @default.
- W4287019225 hasRelatedWork W4285762435 @default.
- W4287019225 isParatext "false" @default.
- W4287019225 isRetracted "false" @default.
- W4287019225 workType "article" @default.