mlsanityData to 14 Oct 2026

Searching for internal symbols underlying deep learning

arXiv 2405.20605 ↗First posted 31 May 2024cs.LG cs.AI cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Deep learning (DL) enables deep neural networks (DNNs) to automatically learn complex tasks or rules from given examples without instructions or guiding principles. As we do not engineer DNNs' functions, it is extremely difficult to diagnose their decisions, and multiple lines of studies proposed to explain the principles of their operations. Notably, one line of studies suggests that DNNs may learn concepts, the high level features that are recognizable to humans. In this study, we extend this line of studies and hypothesize that DNNs can develop abstract codes that can be used to augment DNNs' decision-making. To address this hypothesis, we combine foundation segmentation models and unsupervised learning to extract internal codes and identify potential use of abstract codes to make DL's decision-making more reliable and safer.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 20 of its 36 references that are in the map. First posted on arXiv 31 May 2024.

Builds on

20 of 36 references.

References · 20

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  3. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,733
  4. 2016Densely Connected Convolutional Networkscited by 6,720
  5. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  6. 2015Fast R-CNNcited by 2,933
  7. 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,898
  8. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,451
  9. 2016Understanding intermediate layers using linear classifier probescited by 1,353
  10. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,319
  11. 2016The Mythos of Model Interpretabilitycited by 721
  12. 2017Network Dissection: Quantifying Interpretability of Deep Visual Representationscited by 708
8 more
  1. 2021How to train your ViT? Data, Augmentation, and Regularization in Vision Transformerscited by 569
  2. 2017Adversarial Examples: Attacks and Defenses for Deep Learningcited by 351
  3. 2018Adversarial Attacks and Defences: A Surveycited by 259
  4. 2023In or Out? Fixing ImageNet Out-of-Distribution Detection Evaluationcited by 134
  5. 2018Stop Explaining Black Box Machine Learning Models for High Stakes Decisions and Use Interpretable Models Insteadcited by 40
  6. 2019A Review of Deep Learning with Special Emphasis on Architectures, Applications and Recent Trendscited by 26
  7. 2022Experimental Observations of the Topology of Convolutional Neural Network Activationscited by 10
  8. 2021Explainable Adversarial Attacks in Deep Neural Networks Using Activation Profilescited by 6

Built on by

Not cited yet.