Generative Adversarial Networks: A Survey on Training, Variants, and Applications

Farajzadeh-Zanjani, Maryam; Razavi–Far, Roozbeh; Saif, Mehrdad; Palade, Vasile

doi:10.1007/978-3-030-91390-8_2

Cited by 13 publications

(5 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This involves generating adversarial inputs and including them in the training data, thereby encouraging the model to learn features that are invariant to the adversarial perturbations. While this approach has shown promise, it is not a panacea; adversarial training can sometimes lead to a reduction in accuracy on clean data, as the model may become overly conservative or biased towards the adversarial examples it has encountered during training [15]. The balance between model robustness and performance on clean data is a delicate one, requiring careful calibration of the training process.…”

Section: Training Process and Its Effectsmentioning

confidence: 99%

Deep learning vulnerability analysis against adversarial attacks

Cheng

2024

ACE

View full text Add to dashboard Cite

In the age of artificial intelligence advancements, deep learning models are essential for applications ranging from image recognition to natural language processing. Despite their capabilities, they're vulnerable to adversarial examplesdeliberately modified inputs to cause errors. This paper explores these vulnerabilities, attributing them to the complexity of neural networks, the diversity of training data, and the training methodologies. It demonstrates how these aspects contribute to the models' susceptibility to adversarial attacks. Through case studies and empirical evidence, the paper highlights instances where advanced models were misled, showcasing the challenges in defending against these threats. It also critically evaluates mitigation strategies, including adversarial training and regularization, assessing their efficacy and limitations. The study underlines the importance of developing AI systems that are not only intelligent but also robust against adversarial tactics, aiming to enhance future deep learning models' resilience to such vulnerabilities.

show abstract

Section: Training Process and Its Effectsmentioning

confidence: 99%

Deep learning vulnerability analysis against adversarial attacks

Cheng

2024

ACE

View full text Add to dashboard Cite

show abstract

“…One variation, referred to as the Wasserstein GAN (WGAN) ( 144 , 145 ), introduces a penalty to constrain the gradients of the discriminator’s output, resulting in a more stable and trainable model. While GANs utilize a sigmoid function in the last layer for binary classification, the WGAN approach removes this function to approximate the Wasserstein distance ( 146 ), using Lipschitz discriminators: namely, that for discriminator function

there exists a constant

such that

for any two points

and

in the input space. This ensures that the gradient of the discriminator’s output with respect to its input is bounded by some constant

.…”

Section: Deep Learning Approachesmentioning

confidence: 99%

Quantitative approaches for decoding the specificity of the human T cell repertoire

Ghoreyshi,

George

2023

Front. Immunol.

View full text Add to dashboard Cite

T cell receptor (TCR)-peptide-major histocompatibility complex (pMHC) interactions play a vital role in initiating immune responses against pathogens, and the specificity of TCRpMHC interactions is crucial for developing optimized therapeutic strategies. The advent of high-throughput immunological and structural evaluation of TCR and pMHC has provided an abundance of data for computational approaches that aim to predict favorable TCR-pMHC interactions. Current models are constructed using information on protein sequence, structures, or a combination of both, and utilize a variety of statistical learning-based approaches for identifying the rules governing specificity. This review examines the current theoretical, computational, and deep learning approaches for identifying TCR-pMHC recognition pairs, placing emphasis on each method’s mathematical approach, predictive performance, and limitations.

show abstract

“…This approach can provide an acceptable resilience against evasion attacks [97,110]. While there are different approaches to carry out adversarial training, including the so-called generative adversarial networks [31,28,30,29], non of them are flawless. To begin with, this approach was mainly designed for independent and identically distributed data.…”

Section: Adversarial Trainingmentioning

confidence: 99%

Federated and Transfer Learning: A Survey on Adversaries and Defense Mechanisms

Hallaji,

Razavi-Far,

Saif

2022

Preprint

Self Cite

View full text Add to dashboard Cite

The advent of federated learning has facilitated large-scale data exchange amongst machine learning models while maintaining privacy. Despite its brief history, federated learning is rapidly evolving to make wider use more practical. One of the most significant advancements in this domain is the incorporation of transfer learning into federated learning, which overcomes fundamental constraints of primary federated learning, particularly in terms of security. This chapter performs a comprehensive survey on the intersection of federated and transfer learning from a security point of view. The main goal of this study is to uncover potential vulnerabilities and defense mechanisms that might compromise the privacy and performance of systems that use federated and transfer learning.

show abstract

Generative Adversarial Networks: A Survey on Training, Variants, and Applications

Cited by 13 publications

References 52 publications

Deep learning vulnerability analysis against adversarial attacks

Deep learning vulnerability analysis against adversarial attacks

Quantitative approaches for decoding the specificity of the human T cell repertoire

Federated and Transfer Learning: A Survey on Adversaries and Defense Mechanisms

Contact Info

Product

Resources

About