Efficiently(1) computing the weighted Jaccard similarity has become an active research topic in machine learning and theory. For sparse data, the standard technique is based on the consistent weighed sampling (CWS). F...
详细信息
ISBN:
(纸本)9781577358664
Efficiently(1) computing the weighted Jaccard similarity has become an active research topic in machine learning and theory. For sparse data, the standard technique is based on the consistent weighed sampling (CWS). For dense data, however, methods based on rejection sampling (RS) can be much more efficient. Nevertheless, existing RS methods are still slow for practical purposes. In this paper, we propose to improve RS by a strategy, which we call efficient rejection sampling (ERS), based on "early stopping + densification". We analyze the statistical property of ERS and provide experimental results to compare ERS with RS and other algorithms for hashing weighted Jaccard. The results demonstrate that ERS significantly improves the existing methods for estimating the weighted Jaccard similarity in relatively dense data.
The blockchain is a distributed storage system of digital assets. This decentralized, non-copyable technology stems from universal standard password algorithm and the consensus mechanism of the game theory. The develo...
详细信息
The blockchain is a distributed storage system of digital assets. This decentralized, non-copyable technology stems from universal standard password algorithm and the consensus mechanism of the game theory. The development of quantum computing poses threat to traditional algorithms of blockchain encryption, including symmetric encryption and hash encryption. Focusing on the traditional blockchain consensus mechanism, this paper designs a new blockchain consensus mechanism, based on the stochasticity, irreversibility, and uncertainty of quantum measurement. In the proposed consensus mechanism, complex calculations and intractability mathematical problems are abandoned. In this way, a huge amount of computing resources is saved, less energy is consumed, the time delay is shortened, and the throughput is increased. The proposed quantum consensus mechanism can withstand 51% attacks.
The Molecular Field-Coupled Nanocomputing (FCN) is a computing beyond-CMOS paradigm that encodes the information in the charge distribution of molecules and propagates it through local electrostatic coupling. Notwiths...
详细信息
ISBN:
(纸本)9781665423328
The Molecular Field-Coupled Nanocomputing (FCN) is a computing beyond-CMOS paradigm that encodes the information in the charge distribution of molecules and propagates it through local electrostatic coupling. Notwithstanding the incredibly high potentialities of this technology in the field of high-speed and low-power digital electronics, a molecular prototype has not been produced yet. Indeed, this technology requires nanometric layouts, which are challenging to obtain, slowing down the technology assessment. In this work, we propose a paradigm that bypasses the need for nanometric patterning of molecular devices by organizing the uniform Self-Assembled Monolayer (SAM) into molecular blocks that may store information and be activated independently. The activation of blocks configures the SAM to perform in-memory logic computation. This study demonstrates a reconfigurable molecular standard-cell that maps the basic logic gates (routing, majority voters, inverters), enabling complex digital circuit design. With this paradigm, we move the challenges from the SAM nanopatterning to the clocking system technological feasibility, reducing resolution constraints and favoring the eventual realization of a prototype.
Serverless computing is increasingly seen as a pivot cloud computing paradigm that has great potential to simplify application development while removing the burden of operational tasks from developers. Despite these ...
详细信息
ISBN:
(纸本)9781450386982
Serverless computing is increasingly seen as a pivot cloud computing paradigm that has great potential to simplify application development while removing the burden of operational tasks from developers. Despite these advantages, the use of serverless computing has been limited to few application scenarios exhibiting stateless and parallel executions. In addition, the significant effort and cost associated with re-architecting existing codebase limits the range of these applications and hinder efforts to enhance serverless computing platforms to better suit the needs of current applications. In this paper, we report our experience and observations from migrating four complex and stateful microservice applications (involving 8 programming languages, 5 application frameworks, and 40 application logic services) to Apache OpenWhisk, a widely used serverless computing platform. We highlight a number of patterns and guidelines that facilitate this migration with minimal code changes and practical performance considerations, and imply a path towards further automating this process. We hope our guidelines will help increase the applicability of serverless computing and improve serverless platforms to be more application friendly.
Automated systems that detect the social behavior of deception can enhance human well-being across medical, social work, and legal domains. Labeled datasets to train supervised deception detection models can rarely be...
详细信息
ISBN:
(纸本)9781665431767
Automated systems that detect the social behavior of deception can enhance human well-being across medical, social work, and legal domains. Labeled datasets to train supervised deception detection models can rarely be collected for real-world, high-stakes contexts. To address this challenge, we propose the first unsupervised approach for detecting realworld, high-stakes deception in videos without requiring labels. This paper presents our novel approach for affect-aware unsupervised Deep Belief Networks (DBN) to learn discriminative representations of deceptive and truthful behavior. Drawing on psychology theories that link affect and deception, we experimented with unimodal and multimodal DBN-based approaches trained on facial valence, facial arousal, audio, and visual features. In addition to using facial affect as a feature on which DBN models are trained, we also introduce a DBN training procedure that uses facial affect as an aligner of audio-visual representations. We conducted classification experiments with unsupervised Gaussian Mixture Model clustering to evaluate our approaches. Our best unsupervised approach (trained on facial valence and visual features) achieved an AUC of 80%, outperforming human ability and performing comparably to fully-supervised models. Our results motivate future work on unsupervised, affect-aware computational approaches for detecting deception and other social behaviors in the wild.
Preferences play a key role in computational argumentation in AI, as they reflect various notions of argument strength vital for the representation of argumentation. Within central formal approaches to structured argu...
详细信息
ISBN:
(纸本)9781577358664
Preferences play a key role in computational argumentation in AI, as they reflect various notions of argument strength vital for the representation of argumentation. Within central formal approaches to structured argumentation, preferential approaches are applied by lifting preferences over defeasible elements to rankings over sets of defeasible elements, in order to be able to compare the relative strength of two arguments and their respective defeasible constituents. To overcome the current gap in the scientific landscape, we give in this paper a general study of the critical component of lifting operators in structured argumentation. We survey existing lifting operators scattered in the literature of argumentation theory, social choice, and utility theory, and show fundamental relations and properties of these operators. Extending existing works from argumentation and social choice, we propose a list of postulates for lifting operations, and give a complete picture of (non-)satisfaction for the considered operators. Based on our postulates, we present impossibility results, stating for which sets of postulates there is no hope of satisfaction, and for two main lifting operators presented in structured argumentation, Elitist and Democratic, we give a full characterization in terms of our postulates.
Using combinatorial techniques, we derive a recurrence identity that expresses an exponential power sum with negative powers in terms of another exponential power sum with positive powers. Consequently, we derive a fo...
详细信息
A collapse-free version of quantum theory is introduced to study the role of the projection postulate. We assume "passive" measurements that do not update quantum states while measurement outcomes still occu...
详细信息
We describe three algorithms to determine the stable, semistable, and torus-polystable loci of the GIT quotient of a projective variety by a reductive group. The algorithms are efficient when the group is semisimple. ...
详细信息
The Secure and Trustworthy Computing (SaTC) program within the National Science Foundation (NSF) program serves as the primary instrument for creating novel fundamental science in security and privacy in the United St...
详细信息
The Secure and Trustworthy Computing (SaTC) program within the National Science Foundation (NSF) program serves as the primary instrument for creating novel fundamental science in security and privacy in the United States with broad impacts that influence the world. The program funds research in a vast array of research topics that span technology, theory, policy, law and society. Once a decade, SaTC revisits its mandate and undertakes the task of envisioning a future for cybersecurity research. This comprehensive vision considers the needs and opportunities for research conducted on behalf of the United States, contributing to the nation's advancements in security and privacy. This document serves as the culmination of that effort, providing valuable insights for the future of this critical field. We find ourselves at a unique moment in time, witnessing the rapid advancement of technology that is poised to reshape our society and daily lives. However, along with these transformative capabilities come significant threats from adversaries. Our reliance on online and interconnected systems has never been greater, as the public increasingly depends on technology for personal and professional activities. As such, it is imperative to address the associated risks and vulnerabilities. Recent advancements, such as the remarkable progress in artificial intelligence (AI) and particularly in large language models, underscore the need for heightened attention to security and privacy. The uncertainty surrounding the security and safety of these new capabilities, especially in the complex political and social landscape, emphasizes the urgency for robust research in these areas. By addressing these imperatives, we aim to safeguard the well-being of the people in the United States and around the world. The critical need for advancements in the science of security and privacy is evident. The interconnectedness of systems, the pervasive use of technology, and the potential risks pose
暂无评论