Yes, we GAN: Applying adversarial techniques for autonomous driving

Uřičář, Michal; Křížek, Pavel; Hurych, David; Sobh, Ibrahim; Yogamani, Senthil; Denny, Patrick

doi:10.2352/issn.2470-1173.2019.15.avm-048

Cited by 25 publications

(14 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, in order to create varying adversarial example generators, an implementation of the Fast Gradient Sign Method (FGSM) was chosen [17]. This framework allows for the generation of adversarial patterns through a white box attack.…”

Section: Adversarial Example Generatormentioning

confidence: 99%

“…FGSM needs complete access to the model that is being attacked. This attack type is largely impossible to achieve in the real world against existing self-driving solutions as base-level access is needed to the model, the architecture, and the vehicle that is being attacked [17]. Therefore, this attack is not possible to recreate in the real world; however, it can still show the impact of what changing model architecture does on the effectiveness of the adversarial images.…”

Section: Adversarial Example Generatormentioning

confidence: 99%

See 1 more Smart Citation

Impact of Model Architecture Against Adversarial Example's Effectivity

Karnala

Campbell

2021

J Stud Res

View full text Add to dashboard Cite

The purpose of this study is to gain an understanding of the impact of model architecture on the efficacy of adversarial examples against machine learning systems implemented in self-driving applications. Prior research shows how to create and train against adversarial examples in many use cases; however, there is no definite understanding of how a machine learning model’s architecture affects the efficacy of adversarial examples. Data was collected through an experimental setting involving end-to-end self-driving models trained through behavioral cloning. Three model types were tested based on popular frameworks for machine learning algorithms dealing with images. Results showed a statistically significant difference in the impact of adversarial examples between these models. This means that certain model types and architectures are more susceptible to attacks. Therefore, the conclusion can be made that model architecture does impact the efficacy of adversarial examples; however, this is potentially limited to closed-loop, end-to-end systems in which algorithms make the entire decision. Future research should investigate what specific structure within models causes increased susceptibility to adversarial attacks.

show abstract

Section: Adversarial Example Generatormentioning

confidence: 99%

Section: Adversarial Example Generatormentioning

confidence: 99%

Impact of Model Architecture Against Adversarial Example's Effectivity

Karnala

Campbell

2021

J Stud Res

View full text Add to dashboard Cite

show abstract

“…GANs have been successfully applied to many problems, especially those concerning multimedia information (e.g., images, sound, and video), in science, design, art, games, and other areas [13]. Even complex tasks such as autonomous driving [20] and medical assistance [9] have been addressed using GANs. In fact, medical assistance provides a very interesting application area for GANs, since they can provide new insights to the interpretation process of medical information stored in different media (radiography, ultrasound, etc.…”

Section: Introductionmentioning

confidence: 99%

Parallel/Distributed Generative Adversarial Neural Networks for Data Augmentation of COVID-19 Training Images

Toutouh

Esteban

Nesmachnow

2021

Communications in Computer and Information Science

View full text Add to dashboard Cite

This article presents an approach using parallel/distributed generative adversarial networks for image data augmentation, applied to generate COVID-19 training samples for computational intelligence methods. This is a relevant problem nowadays, considering the recent COVID-19 pandemic. Computational intelligence and learning methods are useful tools to assist physicians in the process of diagnosing diseases and acquire valuable medical knowledge. A specific generative adversarial network approach trained using a co-evolutionary algorithm is implemented, including a three-level parallel approach combining distributed memory and fine-grained parallelization using CPU and GPU. The experimental evaluation of the proposed method was performed on the high performance computing infrastructure provided by National Supercomputing Center, Uruguay. The main experimental results indicate that the proposed model is able to generate accurate images and the 3 × 3 version of the distributed GAN has better robustness properties of its training process, allowing to generate better and more diverse images.

show abstract

“…The generative adversarial network (GAN), first introduced by Ian Goodfellow et al [6], is the most powerful framework for training a generative model. Many generative models based on GANs have been proposed and have shown a tremendous success in applications of various fields, including computer vision [7][8][9]. These models can generate new plausible synthetic samples, which alleviate the problems of data imbalance and lack of diversity in the real dataset, thereby improving the performance degradation caused by skewed class proportions [10][11][12][13][14].…”

Section: Introductionmentioning

confidence: 99%

Multi-Scale Conditional Generative Adversarial Network for Small-Sized Lung Nodules Using Class Activation Region Influence Maximization

et al. 2021

View full text Add to dashboard Cite

Automatic detection and classification of thoracic diseases using deep learning algorithms have many applications supporting radiologists' diagnosis and prognosis. However, in the medical field, the classimbalanced problem is extremely common due to the differences in prevalence among diseases, making it difficult to develop these applications. Many GAN-based methods have been proposed to solve the classimbalance problem on chest X-ray (CXR) data. However, these models have not been trained well for smallsized diseases because it is challenging to extract sufficient information with only a few pixels. In this paper, we propose a novel deep generative model called a class activation region influence maximization conditional generative adversarial network (CARIM-cGAN). The proposed network can control the target disease's presence, location, and size with a controllable conditional mask. We newly introduced class activation region influence maximization (CARIM) loss to maximize the probability of disease occurrence in the bounded region represented by a conditional mask. To demonstrate an enhanced generative performance, we conducted numerous qualitative and quantitative evaluations with the samples generated using a CARIM-cGAN. The results showed that our method has a better performance than other methods. In conclusion, because the CARIM-cGAN can generate high-quality samples based on information on the location and size of the disease, we can contribute to solving problems such as disease classification, -detection, andlocalization, requiring a higher annotation cost.

show abstract

Yes, we GAN: Applying adversarial techniques for autonomous driving

Cited by 25 publications

References 51 publications

Impact of Model Architecture Against Adversarial Example's Effectivity

Impact of Model Architecture Against Adversarial Example's Effectivity

Parallel/Distributed Generative Adversarial Neural Networks for Data Augmentation of COVID-19 Training Images

Multi-Scale Conditional Generative Adversarial Network for Small-Sized Lung Nodules Using Class Activation Region Influence Maximization

Contact Info

Product

Resources

About