Resilient Multiscale Coordination Control against Adversarial Nodes

نویسندگان

چکیده

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Distributed Optimization Under Adversarial Nodes

We investigate the vulnerabilities of consensus-based distributed optimization protocols to nodes that deviate from the prescribed update rule (e.g., due to failures or adversarial attacks). We first characterize certain fundamental limitations on the performance of any distributed optimization algorithm in the presence of adversaries. We then propose a resilient distributed optimization algori...

متن کامل

Safe and Resilient Multi-vehicle Trajectory Planning Under Adversarial Intruder

Provably safe and scalable multi-vehicle trajectory planning is an important and urgent problem. Hamilton-Jacobi (HJ) reachability is an ideal tool for analyzing such safetycritical systems and has been successfully applied to several smallscale problems. However, a direct application of HJ reachability to multi-vehicle trajectory planning is often intractable due to the “curse of dimensionalit...

متن کامل

Defense against Universal Adversarial Perturbations

Recent advances in Deep Learning show the existence of image-agnostic quasi-imperceptible perturbations that when applied to ‘any’ image can fool a state-of-the-art network classifier to change its prediction about the image label. These ‘Universal Adversarial Perturbations’ pose a serious threat to the success of Deep Learning in practice. We present the first dedicated framework to effectivel...

متن کامل

Guarding against Adversarial Domain Shifts

When training a deep neural network for supervised image classification, one can broadly distinguish between two types of latent features of images that will drive the classification: (i) ‘immutable’ or ‘core’ features that are really inherent to the object in question and do not change substantially from one instance of the object to another and (ii) ‘mutable’ or ‘style’ features such as posit...

متن کامل

Certified Defenses against Adversarial Examples

While neural networks have achieved high accuracy on standard image classification benchmarks, their accuracy drops to nearly zero in the presence of small adversarial perturbations to test inputs. Defenses based on regularization and adversarial training have been proposed, but often followed by new, stronger attacks that defeat these defenses. Can we somehow end this arms race? In this work, ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Energies

سال: 2018

ISSN: 1996-1073

DOI: 10.3390/en11071844