mlsanityData to 7 Oct 2026

Generator Born from Classifier

arXiv 2312.02470 ↗First posted 5 Dec 2023cs.LG cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

In this paper, we make a bold attempt toward an ambitious task: given a pre-trained classifier, we aim to reconstruct an image generator, without relying on any data samples. From a black-box perspective, this challenge seems intractable, since it inevitably involves identifying the inverse function for a classifier, which is, by nature, an information extraction process. As such, we resort to leveraging the knowledge encapsulated within the parameters of the neural network. Grounded on the theory of Maximum-Margin Bias of gradient descent, we propose a novel learning paradigm, in which the generator is trained to ensure that the convergence conditions of the network parameters are satisfied over the generated distribution of the samples. Empirical validation from various image generation tasks substantiates the efficacy of our strategy.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 46 of its 61 references that are in the map. First posted on arXiv 5 Dec 2023.

Builds on

46 of 61 references.

References · 46

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
  3. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  4. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,590
  5. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
  6. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  7. 2014Conditional Generative Adversarial Netscited by 2,152
  8. 2016Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Networkcited by 1,898
  9. 2017A Closer Look at Memorization in Deep Networkscited by 936
  10. 2016InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Netscited by 858
  11. 2017The Implicit Bias of Gradient Descent on Separable Datacited by 649
  12. 2023LLM-Pruner: On the Structural Pruning of Large Language Modelscited by 643
34 more
  1. 2018Generative Image Inpainting with Contextual Attentioncited by 578
  2. 2016Conditional Image Synthesis With Auxiliary Classifier GANscited by 562
  3. 2016Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modelingcited by 558
  4. 2017MoCoGAN: Decomposing Motion and Content for Video Generationcited by 523
  5. 2016f-GAN: Training Generative Neural Samplers using Variational Divergence Minimizationcited by 504
  6. 2020What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimationcited by 492
  7. 2019Dreaming to Distill: Data-free Knowledge Transfer via DeepInversioncited by 419
  8. 2019Your Classifier is Secretly an Energy Based Model and You Should Treat it Like Onecited by 402
  9. 2016Generating Videos with Scene Dynamicscited by 376
  10. 2016Adversarial Feature Learningcited by 350
  11. 2019Gradient Descent Maximizes the Margin of Homogeneous Neural Networkscited by 308
  12. 2015Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networkscited by 214
  13. 2023DepGraph: Towards Any Structural Pruningcited by 202
  14. 2023Structural Pruning for Diffusion Modelscited by 193
  15. 2020Directional convergence and alignment in deep learningcited by 172
  16. 2016Semantic Image Inpainting with Deep Generative Modelscited by 156
  17. 2018Regularization Matters: Generalization and Optimization of Neural Nets v.s. their Induced Kernelcited by 131
  18. 2022Reconstructing Training Data from Trained Neural Networkscited by 129
  19. 2018Stochastic subgradient method converges on tame functionscited by 120
  20. 2016Multifaceted Feature Visualization: Uncovering the Different Types of Features Learned By Each Neuron in Deep Neural Networkscited by 106
  21. 2021Gradient Inversion with Generative Image Priorcited by 101
  22. 2016High-Resolution Image Inpainting using Multi-Scale Neural Patch Synthesiscited by 89
  23. 2022Deep Model Reassemblycited by 78
  24. 2017GP-GAN: Towards Realistic High-Resolution Image Blendingcited by 64
  25. 2022Factorizing Knowledge in Neural Networkscited by 59
  26. 2019Lexicographic and Depth-Sensitive Margins in Homogeneous and Non-Homogeneous Deep Modelscited by 53
  27. 2018Improving the Improved Training of Wasserstein GANs: A Consistency Term and Its Dual Effectcited by 49
  28. 2021Exploiting Explanations for Model Inversion Attackscited by 48
  29. 2017LR-GAN: Layered Recursive Generative Adversarial Networks for Image Generationcited by 36
  30. 2022Training invariances and the low-rank phenomenon: beyond linear networkscited by 33
  31. 2022The Asymmetric Maximum Margin Bias of Quasi-Homogeneous Neural Networkscited by 25
  32. 2019Neural Networks Learning and Memorization with (almost) no Over-Parameterizationcited by 23
  33. 2023Distribution Shift Inversion for Out-of-Distribution Predictioncited by 23
  34. 2023Deep Graph Reprogrammingcited by 22

Built on by

Cited by year

Citing papers per year: 2023: 1; 2024: 2; 2025: 1.

4 papers.

  1. 2025VITAL: More Understandable Feature Visualization through Distribution Alignment and Relevant Information Flowcited by 3
  2. 2023Rethinking Model Inversion Attacks With Patch-Wise Reconstructioncited by 0
  3. 2024Deep Support Vectorscited by 0
  4. 2024Slowing Down Forgetting in Continual Learningcited by 0