mlsanityData to 14 Oct 2026

Meta Internal Learning

arXiv 2110.02900 ↗First posted 6 Oct 2021cs.CVcited by 7

Not on the map. Why?

Abstract (arXiv)

Internal learning for single-image generation is a framework, where a generator is trained to produce novel images based on a single image. Since these models are trained on a single image, they are limited in their scale and application. To overcome these issues, we propose a meta-learning approach that enables training over a collection of images, in order to model the internal statistics of the sample image more effectively. In the presented meta-learning approach, a single-image GAN model is generated given an input image, via a convolutional feedforward hypernetwork f. This network is trained over a dataset of images, allowing for feature sharing among different models, and for interpolation in the space of generative models. The generated single-image model contains a hierarchy of multiple generators and discriminators. It is therefore required to train the meta-learner in an adversarial manner, which requires careful design choices that we justify by a theoretical analysis. Our results show that the models obtained are as suitable as single-image GANs for many common image applications, significantly reduce the training time per image without loss in performance, and introduce novel capabilities, such as interpolation and feedforward modeling of novel images.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 27 of its 38 references that are in the map. First posted on arXiv 6 Oct 2021.

Builds on

27 of 38 references.

References · 27

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  3. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  4. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,527
  5. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,611
  6. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,238
  7. 2017Improved Training of Wasserstein GANscited by 2,235
  8. 2017Deep Image Priorcited by 1,098
  9. 2016HyperNetworkscited by 1,027
  10. 2019SinGAN: Learning a Generative Model from a Single Natural Imagecited by 348
  11. 2017The Expressive Power of Neural Networks: A View from the Widthcited by 320
  12. 2017"Zero-Shot" Super-Resolution using Deep Internal Learningcited by 286
15 more
  1. 2019Continual learning with hypernetworkscited by 273
  2. 2017SMASH: One-Shot Model Architecture Search through HyperNetworkscited by 202
  3. 2016Precomputed Real-Time Texture Synthesis with Markovian Generative Adversarial Networkscited by 192
  4. 2018Graph HyperNetworks for Neural Architecture Searchcited by 132
  5. 2017Approximating Continuous Functions by ReLU Nets of Minimal Widthcited by 122
  6. 2016Combining Markov Random Fields and Convolutional Neural Networks for Image Synthesiscited by 102
  7. 2016Learning feed-forward one-shot learnerscited by 93
  8. 2018ResNet with one-neuron hidden layers is a Universal Approximatorcited by 79
  9. 2020Improved Techniques for Training Single-Image GANscited by 78
  10. 2018Stochastic Hyperparameter Optimization through Hypernetworkscited by 73
  11. 2016Depth-Width Tradeoffs in Approximating Natural Functions with Neural Networkscited by 59
  12. 2019Deep Meta Functionals for Shape Representationcited by 53
  13. 2019HyperGAN: A Generative Model for Diverse, Performant Neural Networkscited by 51
  14. 2020Hierarchical Patch VAE-GAN: Generating Diverse Videos from a Single Samplecited by 43
  15. 2020On Infinite-Width Hypernetworkscited by 4

Built on by

Cited by year

Citing papers per year: 2022: 4; 2023: 3.

7 papers.

  1. 2023A Brief Review of Hypernetworks in Deep Learningcited by 102
  2. 2022Towards Diverse and Faithful One-shot Adaption of Generative Adversarial Networkscited by 23
  3. 2023Magnitude Invariant Parametrizations Improve Hypernetwork Learningcited by 14
  4. 2022Neuro-Symbolic Learning: Principles and Applications in Ophthalmologycited by 7
  5. 2022Neural Inverse Kinematicscited by 4
  6. 2022One-Shot Synthesis of Images and Segmentation Maskscited by 4
  7. 2023Exploring the Approximation Capabilities of Multiplicative Neural Networks for Smooth Functionscited by 4