mlsanityData to 14 Oct 2026

Concept Evolution in Deep Learning Training: A Unified Interpretation Framework and Discoveries

arXiv 2203.16475 ↗First posted 30 Mar 2022cs.LG cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

We present ConceptEvo, a unified interpretation framework for deep neural networks (DNNs) that reveals the inception and evolution of learned concepts during training. Our work addresses a critical gap in DNN interpretation research, as existing methods primarily focus on post-training interpretation. ConceptEvo introduces two novel technical contributions: (1) an algorithm that generates a unified semantic space, enabling side-by-side comparison of different models during training, and (2) an algorithm that discovers and quantifies important concept evolutions for class predictions. Through a large-scale human evaluation and quantitative experiments, we demonstrate that ConceptEvo successfully identifies concept evolutions across different models, which are not only comprehensible to humans but also crucial for class predictions. ConceptEvo is applicable to both modern DNN architectures, such as ConvNeXt, and classic DNNs, such as VGGs and InceptionV3.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 41 of its 68 references that are in the map. First posted on arXiv 30 Mar 2022.

Builds on

41 of 68 references.

References · 41

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  3. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  4. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  5. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,733
  6. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
  7. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  8. 2013Efficient Estimation of Word Representations in Vector Spacecited by 4,375
  9. 2022A ConvNet for the 2020scited by 3,579
  10. 2013Distributed Representations of Words and Phrases and their Compositionalitycited by 3,474
  11. 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,898
  12. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,546
29 more
  1. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
  2. 2013Visualizing and Understanding Convolutional Networkscited by 2,290
  3. 2017Understanding Black-box Predictions via Influence Functionscited by 1,741
  4. 2016On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minimacited by 1,352
  5. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,319
  6. 2017Visualizing the Loss Landscape of Neural Netscited by 1,130
  7. 2018A Survey Of Methods For Explaining Black Box Modelscited by 973
  8. 2019On the Convergence of Adam and Beyondcited by 841
  9. 2018This Looks Like That: Deep Learning for Interpretable Image Recognitioncited by 737
  10. 2017Network Dissection: Quantifying Interpretability of Deep Visual Representationscited by 708
  11. 2020Overfitting in adversarially robust deep learningcited by 506
  12. 2015Understanding Neural Networks Through Deep Visualizationcited by 334
  13. 2019On Completeness-aware Concept-Based Explanations in Deep Neural Networkscited by 239
  14. 2018Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learningcited by 207
  15. 2018A Convergence Analysis of Gradient Descent for Deep Linear Neural Networkscited by 184
  16. 2019Explaining Classifiers with Causal Concept Effect (CaCE)cited by 135
  17. 2018Net2Vec: Quantifying and Explaining how Concepts are Encoded by Filters in Deep Neural Networkscited by 129
  18. 2016Resnet in Resnet: Generalizing Residual Architecturescited by 110
  19. 2016Multifaceted Feature Visualization: Uncovering the Different Types of Features Learned By Each Neuron in Deep Neural Networkscited by 106
  20. 2017Interpreting Deep Visual Representations via Network Dissectioncited by 103
  21. 2020Neuron Shapley: Discovering the Responsible Neuronscited by 98
  22. 2019The Dangers of Post-hoc Interpretability: Unjustified Counterfactual Explanationscited by 97
  23. 2015Reducing Overfitting in Deep Networks by Decorrelating Representationscited by 94
  24. 2017Direct-Manipulation Visualization of Deep Networkscited by 28
  25. 2017Examining CNN Representations with respect to Dataset Biascited by 27
  26. 2020Bluff: Interactively Deciphering Adversarial Attacks on Deep Neural Networkscited by 12
  27. 2021NeuroCartography: Scalable Automatic Visual Summarization of Concepts in Deep Neural Networkscited by 12
  28. 2020Explaining with Counter Visual Attributes and Examplescited by 5
  29. 2020InstanceFlow: Visualizing the Evolution of Classifier Confusion on the Instance Levelcited by 5

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 1; 2024: 0; 2025: 1.

3 papers.

  1. 2025Mechanistic understanding and validation of large AI models with SemanticLenscited by 29
  2. 2023Concept-Monitor: Understanding DNN training through individual neuronscited by 2
  3. 2022Generalization Beyond Feature Alignment: Concept Activation-Guided Contrastive Learningcited by 0