Activation Maximization Generative Adversarial Nets

By Zhiming Zhou, Han Cai, Shunlin Rong, Yuxuan Song, Kan Ren, Weinan Zhang, Jun Wang, Yong Yu. 2017

80 citations recorded by the source corpus.

Subfields: Generative Models

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Generative Models

1. Wei Ping
2. Zhifeng Kong
3. Bryan Catanzaro
4. Kexin Zhao
5. Yong Yu
6. Lantao Yu
7. Herman Kamper
8. Matthew Baas

Citations in the corpus

• DiffWave: A Versatile Diffusion Model for Audio Synthesis (2020, methodology): “Table 2: The automatic evaluation metrics (FID, IS, mIS, AM, and NDB /K ), and 5-scale Mean Opinion Score (MOS) with 95% confidence intervals for WaveNet, WaveGAN, and the proposed DiffWave on the unconditional generation task.”
• The Unusual Effectiveness of Averaging in GAN Training (2018, background): “We think this happens because IS is not a proper metric to measure difference between two distributions and has many known flaws such as it does not measure intra-class diversity (Barratt & Sharma, 2018; Zhou et al., 2018; Heusel et al., 2017).”
• Improving MMD-GAN Training with Repulsive Loss Function (2018, background): “It has been widely accepted that the quality of generated samples can be significantly improved by integrating labels (Odena et al. (2017); Miyato & Koyama (2018); Zhou et al. (2018)) or even pseudo-labels generated by k-means method (Grinblat et al. (2017)) in the training of discriminator.”
• Pros and Cons of GAN Evaluation Measures (2018, background): “AM Score.”
• Transferring GANs: generating images from limited data (2018, neutral): “Better correlation with human perception has been found in the widely used Inception Score [39], but recent works have also shown its limitations [49].”
• On the Evaluation of Conditional GANs (2019, background): “Most of these metrics either focus on the separability between generated images and real images [21, 30, 50, 16], compute the distance between distributions [10, 12, 2], assess sample quality and diversity from conditional or marginal distributions [33, 11, 55], measure the similarity between genera”
• Understanding the Effectiveness of Lipschitz Constraint in Training of GANs via Gradient Analysis (2018, background): “Although being very popular [2, 3, 4, 5, 6, 7] and widely-used [8, 9, 10, 11, 12, 13, 14], GANs are notoriously hard to train [15], and the underlying obstacles are currently not fully understood.”
• GAN You Hear Me? Reclaiming Unconditional Speech Synthesis from Diffusion Models (2022, methodology): “Activation maximization (AM) measures generator quality by comparing the KL divergence between the classifier class probabilities from real and generated data, while penalizing high classifier entropy samples produced by the generator [16].”
• PassGAN: A Deep Learning Approach for Password Guessing (2017, background): “[18], there have been several improvements on GANs [48,2,54,62,20,34,46,58,43,3,29,39,7,11,70,66,51,26] [5,42,4,9,25,40,72], where each new paper provides novel improvements in the domain.”