Proxy (statistics)

About: Proxy (statistics) is a research topic. Over the lifetime, 5257 publications have been published within this topic receiving 94504 citations. The topic is also known as: proxy variable & proxy measurement.

Use Privacy in Data-Driven Systems: Theory and Experiments with Machine Learnt Programs

Abstract: This paper presents an approach to formalizing and enforcing a class of use privacy properties in data-driven systems. In contrast to prior work, we focus on use restrictions on proxies (i.e. strong predictors) of protected information types. Our definition relates proxy use to intermediate computations that occur in a program, and identify two essential properties that characterize this behavior: 1) its result is strongly associated with the protected information type in question, and 2) it is likely to causally affect the final output of the program. For a specific instantiation of this definition, we present a program analysis technique that detects instances of proxy use in a model, and provides a witness that identifies which parts of the corresponding program exhibit the behavior. Recognizing that not all instances of proxy use of a protected information type are inappropriate, we make use of a normative judgment oracle that makes this inappropriateness determination for a given witness. Our repair algorithm uses the witness of an inappropriate proxy use to transform the model into one that provably does not exhibit proxy use, while avoiding changes that unduly affect classification accuracy. Using a corpus of social datasets, our evaluation shows that these algorithms are able to detect proxy use instances that would be difficult to find using existing techniques, and subsequently remove them while maintaining acceptable classification performance.

How to Catch when Proxies Lie: Verifying the Physical Locations of Network Proxies with Active Geolocation

Abstract: Internet users worldwide rely on commercial network proxies both to conceal their true location and identity, and to control their apparent location. Their reasons range from mundane to security-critical. Proxy operators offer no proof that their advertised server locations are accurate. IP-to-location databases tend to agree with the advertised locations, but there have been many reports of serious errors in such databases. In this study we estimate the locations of 2269 proxy servers from ping-time measurements to hosts in known locations, combined with AS and network information. These servers are operated by seven proxy services, and, according to the operators, spread over 222 countries and territories. Our measurements show that one-third of them are definitely not located in the advertised countries, and another third might not be. Instead, they are concentrated in countries where server hosting is cheap and reliable (e.g. Czech Republic, Germany, Netherlands, UK, USA). In the process, we address a number of technical challenges with applying active geolocation to proxy servers, which may not be directly pingable, and may restrict the types of packets that can be sent through them, e.g. forbidding traceroute. We also test three geolocation algorithms from previous literature, plus two variations of our own design, at the scale of the whole world.