Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
102 tokens/sec
GPT-4o
59 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

MCL-GAN: Generative Adversarial Networks with Multiple Specialized Discriminators (2107.07260v3)

Published 15 Jul 2021 in cs.LG

Abstract: We propose a framework of generative adversarial networks with multiple discriminators, which collaborate to represent a real dataset more effectively. Our approach facilitates learning a generator consistent with the underlying data distribution based on real images and thus mitigates the chronic mode collapse problem. From the inspiration of multiple choice learning, we guide each discriminator to have expertise in a subset of the entire data and allow the generator to find reasonable correspondences between the latent and real data spaces automatically without extra supervision for training examples. Despite the use of multiple discriminators, the backbone networks are shared across the discriminators and the increase in training cost is marginal. We demonstrate the effectiveness of our algorithm using multiple evaluation metrics in the standard datasets for diverse tasks.

View on arXiv
Definition Search Book Streamline Icon: https://streamlinehq.com
References (52)
  1. Generative Adversarial Nets. In NeurIPS. (2014)
  2. Auto-encoding Variational Bayes. In ICLR. (2014)
  3. Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks. In ICCV. (2017)
  4. Diverse Image-to-Image Translation via Disentangled Representations. In ECCV. (2018)
  5. DRIT++: Diverse Image-to-Image Translation via Disentangled Representations. IJCV (2020)
  6. Semantic Image Inpainting with Deep Generative Models. In CVPR. (2017)
  7. Photo-Realistic Single Image Super-Resolution using a Generative Adversarial Network. In CVPR. (2017)
  8. Multiple Choice Learning: Learning to Produce Multiple Structured Outputs. In NeurIPS. (2012)
  9. Dual Discriminator Generative Adversarial Nets. In NeurIPS. (2017)
  10. Generative Multi-Adversarial Networks. (2017)
  11. On-Line Adaptative Curriculum Learning for GANs. In AAAI. (2019)
  12. Mode Seeking Generative Adversarial Networks for Diverse Image Synthesis. In CVPR. (2019)
  13. Multi-Objective Training of Generative Adversarial Networks with Multiple Discriminators. In ICML. (2019)
  14. Stabilizing GAN Training with Multiple Random Projections. arXiv preprint arXiv:1705.07831 (2017)
  15. Stochastic Multiple Choice Learning for Training Diverse Deep Ensembles. In NeurIPS. (2016)
  16. Photographic Image Synthesis with Cascaded Refinement Networks. In ICCV. (2017)
  17. Learning to Specialize with Knowledge Distillation for Visual Question Answering. In NeurIPS. (2018)
  18. DiverseNet: When One Right Answer is not Enough. In CVPR. (2018)
  19. Diverse Image Synthesis from Semantic Layouts via Conditional IMLE. In ICCV. (2019)
  20. Confident Multiple Choice Learning. In ICML. (2017)
  21. Versatile Multiple Choice Learning and its Application to Vision Computing. In CVPR. (2019)
  22. Least Squares Generative Adversarial Networks. In ICCV. (2017)
  23. Wasserstein Generative Adversarial Networks. In ICML. (2017)
  24. Improved Training of Wasserstein GANs. In NeurIPS. (2017)
  25. PacGAN: The Power of Two Samples in Generative Adversarial Networks. In NeurIPS. (2018)
  26. Energy-based Generative Adversarial Networks. In ICLR. (2017)
  27. Adversarial Feature Learning. In ICLR. (2017)
  28. Adversarially Learned Inference. In ICLR. (2017)
  29. VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learning. In NeurIPS. (2017)
  30. Inclusive GAN: Improving Data and Minority Coverage in Generative Models. In ECCV. (2020)
  31. Normalized Diversification. In CVPR. (2019)
  32. Diversity-Sensitive Conditional Generative Adversarial Networks. In ICLR. (2018)
  33. AdaGAN: Boosting Generative Models. In NeurIPS. (2017)
  34. Multi-Agent Diverse Generative Adversarial Networks. In CVPR. (2018)
  35. MGAN: Training Generative Adversarial Nets with Multiple Generators. In ICLR. (2018)
  36. MEGAN: Mixture of Experts of Generative Adversarial Networks for Multimodal Image Generation. In IJCAI. (2018)
  37. Learning Generative Adversarial Networks from Multiple Data Sources. In IJCAI. (2019)
  38. Geometric GAN. arXiv preprint arXiv:1705.02894 (2017)
  39. Gradient-Based Learning Applied to Document Recognition. IEEE (1998)
  40. Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms. arXiv preprint arXiv:1708.07747 (2017)
  41. Krizhevsky, A.: Learning Multiple Layers of Features from Tiny Images. Technical report (2009)
  42. Deep Learning Face Attributes in the Wild. In ICCV. (2015)
  43. Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks. In ICLR. (2016)
  44. Analyzing and Improving the Image Quality of StyleGAN. In CVPR. (2020)
  45. Diverse Image Generation via Self-Conditioned GANs. In CVPR. (2020)
  46. Assessing Generative Models via Precision and Recall. In NeurIPS. (2018)
  47. GANs Trained by a Two Time-Scale Update Rule Converge to a Nash Equilibrium. In NeurIPS. (2017)
  48. On GANs and GMMs. (2018)
  49. The Unreasonable Effectiveness of Deep Features as a Perceptual Metric. In CVPR. (2018)
  50. StackGAN++: Realistic Image Synthesis with Stacked Generative Adversarial Networks. TPAMI (2018)
  51. The Caltech-UCSD Birds-200-2011 Dataset. Technical Report CNS-TR-2011-001, California Institute of Technology (2011)
  52. Adam: A Method for Stochastic Optimization. In ICLR. (2015)
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (2)
  1. Jinyoung Choi (11 papers)
  2. Bohyung Han (86 papers)
Citations (14)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets