DCGANs generate drainage networks quickly from samples.
problem High computational costs in generating large numbers of drainage networks.
method DCGANs trained with connectivity-informed directional information.
result Connectivity-informed DCGANs outperform other methods in reproducing accurate drainage networks.
DCGANs generate realistic breast masses for mammography.
problem Lack of labelled data and imbalanced datasets in medical imaging.
method DCGANs trained on mammographic images to generate synthetic masses.
result DCGANs improve lesion detection in mammography by ~0.09 F1 score.
Generative adversarial networks (GANs) can implicitly learn rich distributions over images, audio, and data which are hard to model with an explicit likelihood. We present a practical Bayesian formulation for unsupervised and semi-supervised learning with GANs. Within this framework, we use stochastic gradient Hamilton…
NTK reveals order and chaos in DNNs, affecting checkerboard and border artifacts.
problem Checkerboard and border artifacts in DNNs.
method Analysis using Neural Tangent Kernel (NTK) in infinite-width setting.
result Transition between order and chaos regimes affects DNN performance.
Scalable method bounds Lipschitz constant of generative models.
problem Bounding the Lipschitz constant of generative models.
method Layerwise convex approximations using zonotopes.
result Efficient and tight bounds on generative models.
Generative adversarial networks (GAN) are improved to better estimate realness.
problem Estimating realness from multiple perspectives in GANs.
method Developed RealnessGAN, which outputs a distribution of realness.
result RealnessGAN outperforms multiple baselines and generates realistic images at high resolution.
MH-GAN uses a discriminator to improve sampling from a GAN's distribution.
problem Improving sampling from a GAN's implicitly defined distribution.
method Combines Markov chain Monte Carlo and GANs, using a discriminator to wrap the generator.
result MH-GAN samples from the true distribution even when the generator is imperfect.
Perturbative GAN reduces training complexity and improves image quality.
problem Training complexity and image quality in GANs.
method Replaces convolution layers with perturbation layers that add fixed noise masks.
result Higher inception score and faster convergence of generated images.
New system uses microwave radar for hand gesture recognition, improving accuracy and reducing processing time.
problem Limited vision-based hand gesture recognition under dark conditions.
method Deformable deep convolutional generative adversarial network (DCGAN) on Doppler radar signals.
result Recognition rate improved by 10% and testing time reduced by 30%.
Proposes MR-GAN to improve GAN training by respecting real data manifold geometry.
problem Challenges in training GANs, especially mode collapse and poor generalization.
method Introduces manifold regularizer to regularize GAN training.
result Improves GAN performance in terms of generalization, equilibrium, and stability.
CGNNs use wavelets for continuous function generation in infinite-dimensional spaces.
problem Generating continuous functions in infinite-dimensional spaces for applications like inverse problems.
method Inspired by DCGAN, CGNNs use wavelet multiresolution analysis with convolutional and nonlinear layers.
result CGNNs can be injective under certain conditions on filters and nonlinearity, leading to Lipschitz stability estimates.
A GAN-based method diagnoses faults in imbalanced industrial time series data.
problem Fault diagnosis in imbalanced industrial time series data.
method Generative adversarial networks (GAN) combined with a feature extractor.
result Our approach achieves excellent performance in detecting faults.
We propose a new algorithm for training generative adversarial networks that jointly learns latent codes for both identities (e.g. individual humans) and observations (e.g. specific photographs). By fixing the identity portion of the latent codes, we can generate diverse images of the same subject, and by fixing the ob…
Automatically assessing emotional valence in human speech has historically been a difficult task for machine learning algorithms. The subtle changes in the voice of the speaker that are indicative of positive or negative emotional states are often "overshadowed" by voice characteristics relating to emotional intensity …
Generative Adversarial Networks (GANs) excel at creating realistic images with complex models for which maximum likelihood is infeasible. However, the convergence of GAN training has still not been proved. We propose a two time-scale update rule (TTUR) for training GANs with stochastic gradient descent on arbitrary GAN…
Proposes GM Score to evaluate GANs considering diversity, disentanglement, and discriminability.
problem Evaluation of GANs for sample quality and diversity.
method Integrates various factors including intra-class and inter-class diversity, disentanglement, and discriminability metrics.
result Demonstrates improved evaluation of GANs on MNIST dataset.
Survey of GANs and autoencoders, addressing mode collapse and likelihood issues.
problem Addressing mode collapse and likelihood issues in GANs and autoencoders.
method Explains various GAN and autoencoder variants, their applications, and methods to resolve issues.
result Various methods to resolve mode collapse and improve likelihood in GANs and autoencoders.
Improves GAN training stability and quality through a tempered learning process.
problem Training instability and low quality samples in GANs.
method Integrates a 'tempering' module that controls the real data distribution, balancing generator and discriminator.
result Improves quality, stability, and convergence speed across various GAN architectures.
A new GAN method uses Student's t-distribution to generate diverse images with less data.
problem GANs require large datasets and often produce nonsensical results.
method Integrates Student's t-distribution with attention mechanism and dual task discriminator.
result Generates diverse and legible images with limited data.
The study examines issues with latent distributions in generative models and proposes using Cauchy distribution.
problem Issues with latent distributions causing mismatch in sampled regions during linear interpolations.
method Proposed using multidimensional Cauchy distribution and two methods for creating non-linear interpolations.
result Linear interpolations may generate unrealistic data due to the Central Limit Theorem, and Cauchy distribution mitigates this issue.
Improves GAN training with a repulsive loss function.
problem Discourages learning of fine details in data.
method Proposes a repulsive loss function and a bounded Gaussian kernel.
result Significantly improves GAN performance without additional computational cost.
New model improves histopathology classification across magnifications.
problem Robust histopathology classification is difficult due to magnification shift.
method Domain-general model using stable sparse embedding signatures.
result Domain-general model outperformed baseline and GAN augmentation.
We study the impact of input noise dimension on GAN performance.
problem Challenges in GANs like mode collapse and unstable training.
method Quantitative analysis using FID and IS on different image datasets.
result The optimal input noise dimension varies by dataset and architecture.
Generative adversarial networks (GANs) are highly effective unsupervised learning frameworks that can generate very sharp data, even for data such as images with complex, highly multimodal distributions. However GANs are known to be very hard to train, suffering from problems such as mode collapse and disturbing visual…
Spider GAN accelerates GAN training with a new approach.
problem Stable training of Generative adversarial networks (GANs).
method Spider GAN leverages a novel approach to identify closely related datasets (friendly neighborhoods) and uses a new measure (signed inception distance) to accelerate GAN training.
result Spider GAN achieves faster convergence and state-of-the-art FID values with one-fifth of the training iterations.
New GAN loss functions improve image generation quality and stability.
problem Improving the performance of GANs in generating high-quality images.
method Introducing least kth-order GAN (LkGAN) and Rényi-centric GAN loss functions. result The proposed loss functions lead to better image quality and stability.