Results 151 to 160 of about 804,777 (293)

EVICT, NEGLECT, OR INVEST? Community Power and the Politics of Urban Informality Governance in Jakarta

International Journal of Urban and Regional Research, EarlyView.
Abstract Why do some informal neighborhoods receive public investment while others are neglected or evicted? This article addresses the inconsistent governmental responses to informal settlements in Jakarta, Indonesia, during the democratic period. State actions range from violent evictions to tolerance and community‐led improvements.
Kadek Wara Urwasi
wiley   +1 more source

A Gradual Adversarial Training Method for Semantic Segmentation

Remote Sensing
Deep neural networks (DNNs) have achieved great success in various computer vision tasks. However, they are susceptible to artificially designed adversarial perturbations, which limit their deployment in security-critical applications.
Yinkai Zan, Pingping Lu, Tingyu Meng
doaj   +1 more source

Western Balkans as the Frontline of Russian Hybrid Warfare

Global Policy, EarlyView.
ABSTRACT Hybrid warfare (HW) scholarship acknowledges the phenomenon's contextual and temporal specificity, yet its dominant conceptual framing has generated a literature largely centred on identifying and categorising hybrid activities. This focus has left the contextual vulnerabilities that enable hybrid threats (HTs) and shape an adversary's ...
Vesna Bojicic‐Dzelilovic
wiley   +1 more source

The law of AI knowledge distillation

American Business Law Journal, EarlyView.
Abstract Knowledge distillation—the practice of training a compact student model on the outputs of a larger teacher model—has emerged as the fastest‐growing technique for replicating and disseminating frontier artificial intelligence (AI) capabilities.
Taorui Guan
wiley   +1 more source

Of hired guns and ideologues: Why would a law firm ever retain an honest expert witness?

Canadian Journal of Economics/Revue canadienne d'économique, EarlyView.
Abstract Suppose that expert witnesses appearing in litigation are, generically, either honest in their assessment of a fact situation or are advocates who testify in favour of their retaining party, regardless of new information. The type of a witness is known to law firms, who engage with them repeatedly, but not to courts.
Martin Richardson, Jose Rodrigues‐Neto
wiley   +1 more source

Diversity-enhanced reconstruction as plug-in defenders against adversarial perturbations

Frontiers in Artificial Intelligence
Deep learning models are susceptible to adversarial examples. In large-scale deployed services, plug-in defenders efficiently defend against such attacks.
Zeshan Pang, Zeshan Pang, Xuehu Yan, Xuehu Yan, Shasha Guo, Shasha Guo, Yuliang Lu, Yuliang Lu   +7 more
doaj   +1 more source

Struggle with Adversarial Defense? Try Diffusion

CoRR
Adversarial attacks induce misclassification by introducing subtle perturbations. Recently, diffusion models are applied to the image classifiers to improve adversarial robustness through adversarial training or by purifying adversarial noise. However, diffusion-based adversarial training often encounters convergence challenges and high computational ...
Yujie Li, Yanbin Wang, Haitao Xu 0002, Bin Liu, Jianguo Sun, Zhenhao Guo, Wenrui Ma   +6 more
openaire   +3 more sources

Moving Target Defense: Defending against Adversarial Defense

, 2019
: A defense-by-randomization framework is proposed as an effective defense mechanism against different types of adversarial attacks on neural networks. Experiments were conducted by selecting a combination of differently constructed image classification ...

core  

Adversarial-Inspired Backdoor Defense via Bridging Backdoor and Adversarial Attacks

Backdoor attacks and adversarial attacks are two major security threats to deep neural networks (DNNs), with the former one is a training-time data poisoning attack that aims to implant backdoor triggers into models by injecting trigger patterns into ...
Wang, Weijian, Yin, Jia-Li, , Lyhwa, Liu, Ximeng, Lin, Wei   +4 more
core   +1 more source

Family‐Led Decision‐Making Meetings as an Alternative to Initial Child Protection Conferences: Findings From a Realist Evaluation of the Safeguarding Family Group Conferencing Pilot in England

Child &Family Social Work, EarlyView.
ABSTRACT Safeguarding Family Group Conferencing (FGC) is a family‐led decision‐making process offered to some families in England as an alternative to an Initial Child Protection Conference when Children's Services have safeguarding concerns. A realist evaluation was conducted within three local authorities piloting this approach to build understanding
Bekkah Bernheim, Jennie Hayes, Alankrita Singh, Katrina Wyatt, Donald Forrester, Lorna Stabler   +5 more
wiley   +1 more source