-
WiFiCue: Public Wireless Access Security Assessment Tool
Authors:
Jonathan K Adams
Abstract:
Public wireless access points are commonly provided by governments, businesses, schools and other organizations and provide access to the Internet for numerous use cases and can present varying degrees of risk to users. While there are steps that can be taken to mitigate public Wi-Fi risks, ranging from avoidance to the application of end-to-end encryption, application specific encryption, and oth…
▽ More
Public wireless access points are commonly provided by governments, businesses, schools and other organizations and provide access to the Internet for numerous use cases and can present varying degrees of risk to users. While there are steps that can be taken to mitigate public Wi-Fi risks, ranging from avoidance to the application of end-to-end encryption, application specific encryption, and other technologies and tools, these options are not always viable. This paper examines risks associated with Wi-Fi from on a network-by-network perspective. Recommender Systems are presented as part of a proposed mechanism for informing users of the risks of connecting to a specific access point. Implementing prototype architecture for this purpose is examined.
△ Less
Submitted 9 October, 2019;
originally announced October 2019.
-
Neural Input Search for Large Scale Recommendation Models
Authors:
Manas R. Joglekar,
Cong Li,
Jay K. Adams,
Pranav Khaitan,
Quoc V. Le
Abstract:
Recommendation problems with large numbers of discrete items, such as products, webpages, or videos, are ubiquitous in the technology industry. Deep neural networks are being increasingly used for these recommendation problems. These models use embeddings to represent discrete items as continuous vectors, and the vocabulary sizes and embedding dimensions, although heavily influence the model's acc…
▽ More
Recommendation problems with large numbers of discrete items, such as products, webpages, or videos, are ubiquitous in the technology industry. Deep neural networks are being increasingly used for these recommendation problems. These models use embeddings to represent discrete items as continuous vectors, and the vocabulary sizes and embedding dimensions, although heavily influence the model's accuracy, are often manually selected in a heuristical manner. We present Neural Input Search (NIS), a technique for learning the optimal vocabulary sizes and embedding dimensions for categorical features. The goal is to maximize prediction accuracy subject to a constraint on the total memory used by all embeddings. Moreover, we argue that the traditional Single-size Embedding (SE), which uses the same embedding dimension for all values of a feature, suffers from inefficient usage of model capacity and training data. We propose a novel type of embedding, namely Multi-size Embedding (ME), which allows the embedding dimension to vary for different values of the feature. During training we use reinforcement learning to find the optimal vocabulary size for each feature and embedding dimension for each value of the feature. In experiments on two common types of large scale recommendation problems, i.e. retrieval and ranking problems, NIS automatically found better vocabulary and embedding sizes that result in $6.8\%$ and $1.8\%$ relative improvements on Recall@1 and ROC-AUC over manually optimized ones.
△ Less
Submitted 9 July, 2019;
originally announced July 2019.
-
Microwave dynamics of high aspect ratio superconducting nanowires studied using self-resonance
Authors:
Daniel F. Santavicca,
Jesse K. Adams,
Lierd E. Grant,
Adam N. McCaughan,
Karl K. Berggren
Abstract:
We study the microwave impedance of extremely high aspect ratio (length/width ~ 5,000) superconducting niobium nitride nanowires. The nanowires are fabricated in a compact meander geometry that is in series with the center conductor of a 50 ohm coplanar waveguide transmission line. The transmission coefficient of the sample is measured up to 20 GHz. At high frequency, a peak in the transmission co…
▽ More
We study the microwave impedance of extremely high aspect ratio (length/width ~ 5,000) superconducting niobium nitride nanowires. The nanowires are fabricated in a compact meander geometry that is in series with the center conductor of a 50 ohm coplanar waveguide transmission line. The transmission coefficient of the sample is measured up to 20 GHz. At high frequency, a peak in the transmission coefficient is seen. Numerical simulations show that this is a half-wave resonance along the length of the nanowire, where the nanowire acts as a high impedance, slow wave transmission line. This resonance sets the upper frequency limit for these nanowires as inductive elements. Fitting simulations to the measured resonance enables a precise determination of the nanowire's complex sheet impedance at the resonance frequency. The real part is a measure of dissipation, while the imaginary part is dominated by kinetic inductance. We characterize the dependence of the sheet resistance and sheet inductance on both temperature and current and compare the results to recent theoretical predictions for disordered superconductors. These results can aid in the understanding of high frequency devices based on superconducting nanowires. They may also lead to the development of novel superconducting devices such as ultra-compact resonators and slow-wave structures.
△ Less
Submitted 12 July, 2016; v1 submitted 22 February, 2016;
originally announced February 2016.
-
Access Control for Hierarchical Joint-Tenancy
Authors:
Jonathan K. Adams,
Basheer N. Bristow
Abstract:
Basic role based access control [RBAC] provides a mechanism for segregating access privileges based upon a user's hierarchical roles within an organization. This model doesn't scale well when there is tight integration of multiple hierarchies. In a case where there is joint-tenancy and a requirement for different levels of disclosure based upon a user's hierarchy, or in our case, organization or…
▽ More
Basic role based access control [RBAC] provides a mechanism for segregating access privileges based upon a user's hierarchical roles within an organization. This model doesn't scale well when there is tight integration of multiple hierarchies. In a case where there is joint-tenancy and a requirement for different levels of disclosure based upon a user's hierarchy, or in our case, organization or company, basic RBAC requires these hierarchies to be effectively merged. Specific roles that effectively represent both the user's organizations and roles must be translated to fit within the merged hierarchy to be used to control access. Essentially, users from multiple organizations are served from a single role base with roles designed to constrain their access as needed.
Our work proposes, through parameterized roles and privileges, a means for accurately representing both users' roles within their respective hierarchies for providing access to controlled objects. Using this method will reduce the amount of complexity required in terms of the number of roles and privileges. The resulting set of roles, privileges, and objects will make modeling and visualizing the access role hierarchy significantly simplified. This paper will give some background on role based access control, parameterized roles and privileges, and then focus on how RBAC with parameterized roles and privileges can be leveraged as an access control solution for the problems presented by joint tenancy.
△ Less
Submitted 22 March, 2006; v1 submitted 22 March, 2006;
originally announced March 2006.
-
A Service-Centric Approach to a Parameterized RBAC Service
Authors:
Jonathan K. Adams
Abstract:
Significant research has been done in the area of Role Based Access Control [RBAC]. Within this research there has been a thread of work focusing on adding parameters to the role and permissions within RBAC. The primary benefit of parameter support in RBAC comes in the form of a significant increase in specificity in how permissions may be granted. This paper focuses on implementing a parameteri…
▽ More
Significant research has been done in the area of Role Based Access Control [RBAC]. Within this research there has been a thread of work focusing on adding parameters to the role and permissions within RBAC. The primary benefit of parameter support in RBAC comes in the form of a significant increase in specificity in how permissions may be granted. This paper focuses on implementing a parameterized implementation based heavily upon existing standards.
△ Less
Submitted 8 March, 2006;
originally announced March 2006.
