Ian Goldberg

Rob Jansen, Justin Tracey, Ian Goldberg

Tor is a popular low-latency anonymous communication system that focuses on usability and performance: a faster network will attract more users, which in turn will improve the anonymity of everyone using the system. The standard practice for previous research attempting to enhance Tor performance is to draw conclusions from the observed results of a single simulation for standard Tor and for each research variant. But because the simulations are run in sampled Tor networks, it is possible that sampling error alone could cause the observed effects. Therefore, we call into question the practical meaning of any conclusions that are drawn without considering the statistical significance of the reported results. In this paper, we build foundations upon which we improve the Tor experimental method. First, we present a new Tor network modeling methodology that produces more representative Tor networks as well as new and improved experimentation tools that run Tor simulations faster and at a larger scale than was previously possible. We showcase these contributions by running simulations with 6,489 relays and 792k simultaneously active users, the largest known Tor network simulations and the first at a network scale of 100%. Second, we present new statistical methodologies through which we: (i) show that running multiple simulations in independently sampled networks is necessary in order to produce informative results; and (ii) show how to use the results from multiple simulations to conduct sound statistical inference. We present a case study using 420 simulations to demonstrate how to apply our methodologies to a concrete set of Tor experiments and how to analyze the results.

Thomas Humphries, Simon Oya, Lindsey Tulloch et al.

Training machine learning models on privacy-sensitive data has become a popular practice, driving innovation in ever-expanding fields. This has opened the door to new attacks that can have serious privacy implications. One such attack, the Membership Inference Attack (MIA), exposes whether or not a particular data point was used to train a model. A growing body of literature uses Differentially Private (DP) training algorithms as a defence against such attacks. However, these works evaluate the defence under the restrictive assumption that all members of the training set, as well as non-members, are independent and identically distributed. This assumption does not hold for many real-world use cases in the literature. Motivated by this, we evaluate membership inference with statistical dependencies among samples and explain why DP does not provide meaningful protection (the privacy parameter $ε$ scales with the training set size $n$) in this more general case. We conduct a series of empirical evaluations with off-the-shelf MIAs using training sets built from real-world data showing different types of dependencies among samples. Our results reveal that training set dependencies can severely increase the performance of MIAs, and therefore assuming that data samples are statistically independent can significantly underestimate the performance of MIAs.

Matthew Lakier, Dimcho Karakashev, Yixin Wang et al.

Smartphones store a significant amount of personal and private information, and are playing an increasingly important role in people's lives. It is important for authentication techniques to be more resistant against two known attacks called shoulder surfing and smudge attacks. In this work, we propose a new technique called 3D Pattern. Our 3D Pattern technique takes advantage of a new input paradigm called pre-touch, which could soon allow smartphones to sense a user's finger position at some distance from the screen. We implement the technique and evaluate it in a pilot study (n=6) by comparing it to PIN and pattern locks. Our results show that although our prototype takes about 8 seconds to authenticate, it is immune to smudge attacks and promises to be more resistant to shoulder surfing.

Stefanie Roos, Pedro Moreno-Sanchez, Aniket Kate et al.

Path-based transaction (PBT) networks, which settle payments from one user to another via a path of intermediaries, are a growing area of research. They overcome the scalability and privacy issues in cryptocurrencies like Bitcoin and Ethereum by replacing expensive and slow on-chain blockchain operations with inexpensive and fast off-chain transfers. In the form of credit networks such as Ripple and Stellar, they also enable low-price real-time gross settlements across different currencies. For example, SilentWhsipers is a recently proposed fully distributed credit network relying on path-based transactions for secure and in particular private payments without a public ledger. At the core of a decentralized PBT network is a routing algorithm that discovers transaction paths between payer and payee. During the last year, a number of routing algorithms have been proposed. However, the existing ad hoc efforts lack either efficiency or privacy. In this work, we first identify several efficiency concerns in SilentWhsipers. Armed with this knowledge, we design and evaluate SpeedyMurmurs, a novel routing algorithm for decentralized PBT networks using efficient and flexible embedding-based path discovery and on-demand efficient stabilization to handle the dynamics of a PBT network. Our simulation study, based on real-world data from the currently deployed Ripple credit network, indicates that SpeedyMurmurs reduces the overhead of stabilization by up to two orders of magnitude and the overhead of routing a transaction by more than a factor of two. Furthermore, using SpeedyMurmurs maintains at least the same success ratio as decentralized landmark routing, while providing lower delays. Finally, SpeedyMurmurs achieves key privacy goals for routing in PBT networks.

Navid Nasr Esfahani, Ian Goldberg, Douglas R. Stinson

A $(t, s, v)$-all-or-nothing transform is a bijective mapping defined on $s$-tuples over an alphabet of size $v$, which satisfies the condition that the values of any $t$ input co-ordinates are completely undetermined, given only the values of any $s-t$ output co-ordinates. The main question we address in this paper is: for which choices of parameters does a $(t, s, v)$-all-or-nothing transform (AONT) exist? More specifically, if we fix $t$ and $v$, we want to determine the maximum integer $s$ such that a $(t, s, v)$-AONT exists. We mainly concentrate on the case $t=2$ for arbitrary values of $v$, where we obtain various necessary as well as sufficient conditions for existence of these objects. We consider both linear and general (linear or nonlinear) AONT. We also show some connections between AONT, orthogonal arrays and resilient functions.

Raphael R. Toledo, George Danezis, Ian Goldberg

Private Information Retrieval (PIR), despite being well studied, is computationally costly and arduous to scale. We explore lower-cost relaxations of information-theoretic PIR, based on dummy queries, sparse vectors, and compositions with an anonymity system. We prove the security of each scheme using a flexible differentially private definition for private queries that can capture notions of imperfect privacy. We show that basic schemes are weak, but some of them can be made arbitrarily safe by composing them with large anonymity systems.

Tariq Elahi, Steven J. Murdoch, Ian Goldberg

This paper argues that one of the most important decisions in designing and deploying censorship resistance systems is whether one set of system options should be selected (the best), or whether there should be several sets of good ones. We model the problem of choosing these options as a cat-and-mouse game and show that the best strategy depends on the value the censor associates with total system censorship versus partial, and the tolerance of false positives. If the censor has a low tolerance to false positives then choosing one censorship resistance system is best. Otherwise choosing several systems is the better choice, but the way traffic should be distributed over the systems depends on the tolerance of the censor to false negatives. We demonstrate that establishing the censor's utility function is critical to discovering the best strategy for censorship resistance.