Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

The author systematically examines one of the important issues of information — establishing the market price. He introduces the concept of «search» — where a buyer wanting to get a better price, is forced to question sellers. The article deals with various aspects of finding the necessary information.

Economics of Information

National Institute of Standards and Technology における超伝導研究及び生活

Today’s artificial intelligence still faces two major challenges. One is that, in most industries, data exists in the form of isolated islands. The other is the strengthening of data privacy and security. We propose a possible solution to these challenges: secure federated learning. Beyond the federated-learning framework first proposed by Google in 2016, we introduce a comprehensive secure federated-learning framework, which includes horizontal federated learning, vertical federated learning, and federated transfer learning. We provide definitions, architectures, and applications for the federated-learning framework, and provide a comprehensive survey of existing works on this subject. In addition, we propose building data networks among organizations based on federated mechanisms as an effective solution to allowing knowledge to be shared without compromising user privacy.

Federated Machine Learning: Concept and Applications

The objective of this paper is to give a comprehensive introduction to applied cryptography with an engineer or computer scientist in mind. The emphasis is on the knowledge needed to create practical systems which supports integrity, confidentiality, or authenticity. Topics covered includes an introduction to the concepts in cryptography, attacks against cryptographic systems, key use and handling, random bit generation, encryption modes, and message authentication codes. Recommendations on algorithms and further reading is given in the end of the paper. This paper should make the reader able to build, understand and evaluate system descriptions and designs based on the cryptographic components described in the paper.

[서평]「Applied Cryptography」

Textual deception constitutes a major problem for online security. Many studies have argued that deceptiveness leaves traces in writing style, which could be detected using text classification techniques. By conducting an extensive literature review of existing empirical work, we demonstrate that while certain linguistic features have been indicative of deception in certain corpora, they fail to generalize across divergent semantic domains. We suggest that deceptiveness as such leaves no content-invariant stylistic trace, and textual similarity measures provide superior means of classifying texts as potentially deceptive. Additionally, we discuss forms of deception beyond semantic content, focusing on hiding author identity by writing style obfuscation. Surveying the literature on both author identification and obfuscation techniques, we conclude that current style transformation methods fail to achieve reliable obfuscation while simultaneously ensuring semantic faithfulness to the original text. We propose that future work in style transformation should pay particular attention to disallowing semantically drastic changes.

Text Analysis in Adversarial Settings: Does Deception Leave a Stylistic Trace?

The ability to access remote file storage from mobile devices enables a number of new use cases for storing and sharing data. We describe the design and implementation of a remote storage client framework on Symbian OS, the leading smart phone OS on the market. Our work is inspired and informed by previous work like Coda M. Satyanarayanan (1989). We describe why Coda cannot be used directly in our target scenarios and environments. We then describe how we adapted the Coda concepts to suit our needs. The advanced features supported by the framework include disconnected operation with whole-file caching and immediate file access (adapting whole-file caching principle to multimedia-centric smart phones). Using this framework, we have implemented Symbian OS remote file systems based on WebDAV and FTP

/pdf/remote-storage-for-mobile-devices-1w36niwnr0.pdf

Remote Storage for Mobile Devices

Initialization of security in a system must correctly capture the underlying prior context of trust assumptions. Technical security mechanisms like security protocols and cryptographic algorithms cannot be effective unless the initialization is done properly. We identify three sources of difficulty in initializing security: (1) basing initialization on real-world identities is expensive, (2) some usage scenarios require new types of indexical prior contexts, and (3) in certain systems, there may not be any real world prior context to link to. We provide an overview of these difficulties and how some recent research work is attempting to address them

Issues in initializing security

A method for authenticating communicating parties is disclosed. In the method biometric information associated with a first party is generated based on a recording of the first party presenting a predefined input parameter. Said biometric information may then be transmitted to a second party. Authenticity of a security parameter associated with the first party can then be verified based on said biometric information.

Authenticating security parameters

A personal device is to be connected to a terminal and being equipped with a computerized method for establishing a trustworthy connection between a user via said device and the terminal which is connected to and authenticatable by at least one server which is authenticatable by the device. In the case the device is coupled to the terminal, a first authentication step is initiatable during which the terminal authenticates itself to the server, upon success of which, between the server and the terminal, a first authenticated trusted connection is established. Subsequently a second authentication step is initiatable during which via the established first authenticated trusted connection the server authenticates itself to the device, upon success of which, between the server and the device, a second authenticated trusted connection is established. Subsequently at the device during a first messaging step, from the server via the established second authenticated trusted connection, a terminal authenticity message is receivable confirming the established authenticity of the terminal. Subsequently during a second messaging step an authenticity output message is communicatable from the device to the user via a device output of the device and/or via a terminal output of the terminal.

Nadarajah Asokan

Papers

Text Analysis in Adversarial Settings: Does Deception Leave a Stylistic Trace?

Remote Storage for Mobile Devices

Issues in initializing security

Authenticating security parameters

Method and system for establishing a trustworthy connection between a user and a terminal