On the Value of Labeled Data and Symbolic Methods for Hidden Neuron Activation Analysis
A major challenge in Explainable AI is in correctly interpreting activations of hidden neurons: accurate interpretations would help answer the question of what a deep learning system internally detects as relevant in the input, demystifying the otherwise black-box nature of deep learning systems. Th...
Saved in:
Published in | arXiv.org |
---|---|
Main Authors | , , , , , |
Format | Paper |
Language | English |
Published |
Ithaca
Cornell University Library, arXiv.org
21.04.2024
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | A major challenge in Explainable AI is in correctly interpreting activations of hidden neurons: accurate interpretations would help answer the question of what a deep learning system internally detects as relevant in the input, demystifying the otherwise black-box nature of deep learning systems. The state of the art indicates that hidden node activations can, in some cases, be interpretable in a way that makes sense to humans, but systematic automated methods that would be able to hypothesize and verify interpretations of hidden neuron activations are underexplored. This is particularly the case for approaches that can both draw explanations from substantial background knowledge, and that are based on inherently explainable (symbolic) methods. In this paper, we introduce a novel model-agnostic post-hoc Explainable AI method demonstrating that it provides meaningful interpretations. Our approach is based on using a Wikipedia-derived concept hierarchy with approximately 2 million classes as background knowledge, and utilizes OWL-reasoning-based Concept Induction for explanation generation. Additionally, we explore and compare the capabilities of off-the-shelf pre-trained multimodal-based explainable methods. Our results indicate that our approach can automatically attach meaningful class expressions as explanations to individual neurons in the dense layer of a Convolutional Neural Network. Evaluation through statistical analysis and degree of concept activation in the hidden layer show that our method provides a competitive edge in both quantitative and qualitative aspects compared to prior work. |
---|---|
AbstractList | A major challenge in Explainable AI is in correctly interpreting activations of hidden neurons: accurate interpretations would help answer the question of what a deep learning system internally detects as relevant in the input, demystifying the otherwise black-box nature of deep learning systems. The state of the art indicates that hidden node activations can, in some cases, be interpretable in a way that makes sense to humans, but systematic automated methods that would be able to hypothesize and verify interpretations of hidden neuron activations are underexplored. This is particularly the case for approaches that can both draw explanations from substantial background knowledge, and that are based on inherently explainable (symbolic) methods. In this paper, we introduce a novel model-agnostic post-hoc Explainable AI method demonstrating that it provides meaningful interpretations. Our approach is based on using a Wikipedia-derived concept hierarchy with approximately 2 million classes as background knowledge, and utilizes OWL-reasoning-based Concept Induction for explanation generation. Additionally, we explore and compare the capabilities of off-the-shelf pre-trained multimodal-based explainable methods. Our results indicate that our approach can automatically attach meaningful class expressions as explanations to individual neurons in the dense layer of a Convolutional Neural Network. Evaluation through statistical analysis and degree of concept activation in the hidden layer show that our method provides a competitive edge in both quantitative and qualitative aspects compared to prior work. |
Author | Rushrukh Rayan Barua, Adrita Md Kamruzzaman Sarker Dalal, Abhilekha Vasserman, Eugene Y Hitzler, Pascal |
Author_xml | – sequence: 1 givenname: Abhilekha surname: Dalal fullname: Dalal, Abhilekha – sequence: 2 fullname: Rushrukh Rayan – sequence: 3 givenname: Adrita surname: Barua fullname: Barua, Adrita – sequence: 4 givenname: Eugene surname: Vasserman middlename: Y fullname: Vasserman, Eugene Y – sequence: 5 fullname: Md Kamruzzaman Sarker – sequence: 6 givenname: Pascal surname: Hitzler fullname: Hitzler, Pascal |
BookMark | eNqNyrsKwjAUgOEgClbtOxxwLsSkalfxgoOXQdGxxOYUU2KONqng26vgAzj9__D1WNuRwxaLhJSjJEuF6LLY-4pzLiZTMR7LiJ33DsIV4aRsg0AlbNQFLWpYqKBAOQ2H1-1C1hSwxXAl7aGkGtZGa3Sww6YmB7MimKcK5rtO2Zc3fsA6pbIe41_7bLhaHufr5F7To0Ef8oqa-oN9Lnma8umI80z-p94CqEIM |
ContentType | Paper |
Copyright | 2024. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
Copyright_xml | – notice: 2024. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
DBID | 8FE 8FG ABJCF ABUWG AFKRA AZQEC BENPR BGLVJ CCPQU DWQXO HCIFZ L6V M7S PIMPY PQEST PQQKQ PQUKI PRINS PTHSS |
DatabaseName | ProQuest SciTech Collection ProQuest Technology Collection Materials Science & Engineering Collection ProQuest Central (Alumni Edition) ProQuest Central ProQuest Central Essentials AUTh Library subscriptions: ProQuest Central Technology Collection ProQuest One Community College ProQuest Central SciTech Premium Collection (Proquest) (PQ_SDU_P3) ProQuest Engineering Collection Engineering Database Publicly Available Content Database ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Academic ProQuest One Academic UKI Edition ProQuest Central China Engineering Collection |
DatabaseTitle | Publicly Available Content Database Engineering Database Technology Collection ProQuest Central Essentials ProQuest One Academic Eastern Edition ProQuest Central (Alumni Edition) SciTech Premium Collection ProQuest One Community College ProQuest Technology Collection ProQuest SciTech Collection ProQuest Central China ProQuest Central ProQuest Engineering Collection ProQuest One Academic UKI Edition ProQuest Central Korea Materials Science & Engineering Collection ProQuest One Academic Engineering Collection |
DatabaseTitleList | Publicly Available Content Database |
Database_xml | – sequence: 1 dbid: 8FG name: ProQuest Technology Collection url: https://search.proquest.com/technologycollection1 sourceTypes: Aggregation Database |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Physics |
EISSN | 2331-8422 |
Genre | Working Paper/Pre-Print |
GroupedDBID | 8FE 8FG ABJCF ABUWG AFKRA ALMA_UNASSIGNED_HOLDINGS AZQEC BENPR BGLVJ CCPQU DWQXO FRJ HCIFZ L6V M7S M~E PIMPY PQEST PQQKQ PQUKI PRINS PTHSS |
ID | FETCH-proquest_journals_30440710083 |
IEDL.DBID | 8FG |
IngestDate | Thu Oct 10 20:31:16 EDT 2024 |
IsOpenAccess | true |
IsPeerReviewed | false |
IsScholarly | false |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-proquest_journals_30440710083 |
OpenAccessLink | https://www.proquest.com/docview/3044071008?pq-origsite=%requestingapplication% |
PQID | 3044071008 |
PQPubID | 2050157 |
ParticipantIDs | proquest_journals_3044071008 |
PublicationCentury | 2000 |
PublicationDate | 20240421 |
PublicationDateYYYYMMDD | 2024-04-21 |
PublicationDate_xml | – month: 04 year: 2024 text: 20240421 day: 21 |
PublicationDecade | 2020 |
PublicationPlace | Ithaca |
PublicationPlace_xml | – name: Ithaca |
PublicationTitle | arXiv.org |
PublicationYear | 2024 |
Publisher | Cornell University Library, arXiv.org |
Publisher_xml | – name: Cornell University Library, arXiv.org |
SSID | ssj0002672553 |
Score | 3.54259 |
SecondaryResourceType | preprint |
Snippet | A major challenge in Explainable AI is in correctly interpreting activations of hidden neurons: accurate interpretations would help answer the question of what... |
SourceID | proquest |
SourceType | Aggregation Database |
SubjectTerms | Activation analysis Artificial neural networks Deep learning Explainable artificial intelligence Machine learning Neurons Statistical analysis |
Title | On the Value of Labeled Data and Symbolic Methods for Hidden Neuron Activation Analysis |
URI | https://www.proquest.com/docview/3044071008 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV07T8MwED5BIyQ2nuJRqpNgjUhsN2kmxCMlQqRUPLtVjuNMkJYmHVj47ZytBAakDh4sS5ZtWffdnX3fB3DGAmXqJSO3kBSiiJyFrgy15_pBFHIlvEIIU42cjoLkRdxN-pMm4VY13ypbm2gNdT5TJkd-zo02sqWiuZh_ukY1yryuNhIa6-D4hgnPVIoPb39zLCwIyWPm_8ysxY7hFjhjOdeLbVjT5Q5s2C-XqtqFt4cSyfvCV_m-1Dgr8F5mhAE53shaIgX4-PT1kRnaXkytzHOF5GBiYjg_SrSkGiVeqlaeDFt6kT04HcbP14nbrmba3Jdq-rc7vg8dCvz1ASAhR59akbNoIJSnByLyIpZFGfdzrwj5IXRXzXS0evgYNhkBtHkZYX4XOvViqU8IYOusZ0-xB85VPBo_Ui_9jn8A1L2EIw |
link.rule.ids | 786,790,12792,21416,33406,33777,43633,43838 |
linkProvider | ProQuest |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV27TsMwFL2CRAg2nqJQ4EqwRiSOm8eEeLQKkIQKCnSLHMeZIC1NOvD32FYCA1IHT5Ys27LuuS-fA3BBPK7-S4ZWyWSIQgviW8wXtuV4oe9yapeUqt_ISepFr_RhOpi2Cbe6bavsbKI21MWMqxz5pau0kTUVzdX8y1KqUaq62kporIOpKDcDA8ybYTp-_s2yEM-XPrP7z9Bq9Bhtgzlmc7HYgTVR7cKGbrrk9R68P1Uo_S98Yx9LgbMSY5ZLFCjwjjUMZYiPL9-fuSLuxUQLPdcoXUyMFOtHhZpWo8Jr3gmUYUcwsg_no-HkNrK63WTti6mzv_O5B2DI0F8cAkrsGMhRFiQMKLdFQEM7JHmYu05hl77bg_6qlY5WT5_BZjRJ4iy-Tx-PYYtIuFZ1EuL0wWgWS3Ei4bbJT9s7_QEruYWv |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=On+the+Value+of+Labeled+Data+and+Symbolic+Methods+for+Hidden+Neuron+Activation+Analysis&rft.jtitle=arXiv.org&rft.au=Dalal%2C+Abhilekha&rft.au=Rushrukh+Rayan&rft.au=Barua%2C+Adrita&rft.au=Vasserman%2C+Eugene+Y&rft.date=2024-04-21&rft.pub=Cornell+University+Library%2C+arXiv.org&rft.eissn=2331-8422 |