arXiv subjects

Benjamin I. P. Rubinstein

Publications and source records attributed to Benjamin I. P. Rubinstein.

At least 73 records · Page 4Linked to original sources

arXiv2018

Options for encoding names for data linking at the Australian Bureau of Statistics

Publicly, ABS has said it would use a cryptographic hash function to convert names collected in the 2016 Census of Population and Housing into an unrecognisable value in a way that is not reversible. In 2016, the ABS engaged the University of Melbourne to provide expert advice on cryptographic hash functions to meet this objective. For complex unit-record level data, including Census data, auxiliary data can be often be used to link individual records, even without names. This is the basis of ABS's existing bronze linking. This means that records can probably be re-identified without the encoded name anyway. Protection against re-identification depends on good processes within ABS. The undertaking on the encoding of names should therefore be considered in the full context of auxiliary data and ABS processes. There are several reasonable interpretations: 1. That the encoding cannot be reversed except with a secret key held by ABS. This is the property achieved by encryption (Option 1), if properly implemented; 2. That the encoding, taken alone without auxiliary data, cannot be reversed to a single value. This is the property achieved by lossy encoding (Option 2), if properly implemented; 3. That the encoding doesn't make re-identification easier, or increase the number of records that can be re-identified, except with a secret key held by ABS. This is the property achieved by HMAC-based linkage key derivation using subsets of attributes (Option 3), if properly implemented. We explain and compare the privacy and accuracy guarantees of five possible approaches. Options 4 and 5 investigate more sophisticated options for future data linking. We also explain how some commonly-advocated techniques can be reversed, and hence should not be used.

cs.CR↗

arXiv2018

Differentially-Private Counting of Users' Spatial Regions

Mining of spatial data is an enabling technology for mobile services, Internet-connected cars, and the Internet of Things. But the very distinctiveness of spatial data that drives utility, can cost user privacy. Past work has focused upon points and trajectories for differentially-private release. In this work, we continue the tradition of privacy-preserving spatial analytics, focusing not on point or path data, but on planar spatial regions. Such data represents the area of a user's most frequent visitation---such as "around home and nearby shops". Specifically we consider the differentially-private release of data structures that support range queries for counting users' spatial regions. Counting planar regions leads to unique challenges not faced in existing work. A user's spatial region that straddles multiple data structure cells can lead to duplicate counting at query time. We provably avoid this pitfall by leveraging the Euler characteristic for the first time with differential privacy. To address the increased sensitivity of range queries to spatial region data, we calibrate privacy-preserving noise using bounded user region size and a constrained inference that uses robust least absolute deviations. Our novel constrained inference reduces noise and promotes covertness by (privately) imposing consistency. We provide a full end-to-end theoretical analysis of both differential privacy and high-probability utility for our approach using concentration bounds. A comprehensive experimental study on several real-world datasets establishes practical validity.

cs.DB↗

arXiv2017

Health Data in an Open World

With the aim of informing sound policy about data sharing and privacy, we describe successful re-identification of patients in an Australian de-identified open health dataset. As in prior studies of similar datasets, a few mundane facts often suffice to isolate an individual. Some people can be identified by name based on publicly available information. Decreasing the precision of the unit-record level data, or perturbing it statistically, makes re-identification gradually harder at a substantial cost to utility. We also examine the value of related datasets in improving the accuracy and confidence of re-identification. Our re-identifications were performed on a 10% sample dataset, but a related open Australian dataset allows us to infer with high confidence that some individuals in the sample have been correctly re-identified. Finally, we examine the combination of the open datasets with some commercial datasets that are known to exist but are not in our possession. We show that they would further increase the ease of re-identification.

cs.CY↗

arXiv2017

Vulnerabilities in the use of similarity tables in combination with pseudonymisation to preserve data privacy in the UK Office for National Statistics' Privacy-Preserving Record Linkage

In the course of a survey of privacy-preserving record linkage, we reviewed the approach taken by the UK Office for National Statistics (ONS) as described in their series of reports "Beyond 2011". Our review identifies a number of matters of concern. Some of the issues discovered are sufficiently severe to present a risk to privacy.

cs.CR↗

arXiv2017

In Search of an Entity Resolution OASIS: Optimal Asymptotic Sequential Importance Sampling

Entity resolution (ER) presents unique challenges for evaluation methodology. While crowdsourcing platforms acquire ground truth, sound approaches to sampling must drive labelling efforts. In ER, extreme class imbalance between matching and non-matching records can lead to enormous labelling requirements when seeking statistically consistent estimates for rigorous evaluation. This paper addresses this important challenge with the OASIS algorithm: a sampler and F-measure estimator for ER evaluation. OASIS draws samples from a (biased) instrumental distribution, chosen to ensure estimators with optimal asymptotic variance. As new labels are collected OASIS updates this instrumental distribution via a Bayesian latent variable model of the annotator oracle, to quickly focus on unlabelled items providing more information. We prove that resulting estimates of F-measure, precision, recall converge to the true population values. Thorough comparisons of sampling methods on a variety of ER datasets demonstrate significant labelling reductions of up to 83% without loss to estimate accuracy.

cs.LG↗

arXiv2017

Pain-Free Random Differential Privacy with Sensitivity Sampling

Popular approaches to differential privacy, such as the Laplace and exponential mechanisms, calibrate randomised smoothing through global sensitivity of the target non-private function. Bounding such sensitivity is often a prohibitively complex analytic calculation. As an alternative, we propose a straightforward sampler for estimating sensitivity of non-private mechanisms. Since our sensitivity estimates hold with high probability, any mechanism that would be $(ε,δ)$-differentially private under bounded global sensitivity automatically achieves $(ε,δ,γ)$-random differential privacy (Hall et al., 2012), without any target-specific calculations required. We demonstrate on worked example learners how our usable approach adopts a naturally-relaxed privacy guarantee, while achieving more accurate releases even for non-private functions that are black-box computer programs.

cs.LG↗

arXiv2017

Adequacy of the Gradient-Descent Method for Classifier Evasion Attacks

Despite the wide use of machine learning in adversarial settings including computer security, recent studies have demonstrated vulnerabilities to evasion attacks---carefully crafted adversarial samples that closely resemble legitimate instances, but cause misclassification. In this paper, we examine the adequacy of the leading approach to generating adversarial samples---the gradient descent approach. In particular (1) we perform extensive experiments on three datasets, MNIST, USPS and Spambase, in order to analyse the effectiveness of the gradient-descent method against non-linear support vector machines, and conclude that carefully reduced kernel smoothness can significantly increase robustness to the attack; (2) we demonstrate that separated inter-class support vectors lead to more secure models, and propose a quantity similar to margin that can efficiently predict potential susceptibility to gradient-descent attacks, before the attack is launched; and (3) we design a new adversarial sample construction algorithm based on optimising the multiplicative ratio of class decision functions.

cs.CR↗

arXiv2017

Privacy Assessment of De-identified Opal Data: A report for Transport for NSW

We consider the privacy implications of public release of a de-identified dataset of Opal card transactions. The data was recently published at https://opendata.transport.nsw.gov.au/dataset/opal-tap-on-and-tap-off. It consists of tap-on and tap-off counts for NSW's four modes of public transport, collected over two separate week-long periods. The data has been further treated to improve privacy by removing small counts, aggregating some stops and routes, and perturbing the counts. This is a summary of our findings.

cs.CR↗

arXiv2017

TopicResponse: A Marriage of Topic Modelling and Rasch Modelling for Automatic Measurement in MOOCs

This paper explores the suitability of using automatically discovered topics from MOOC discussion forums for modelling students' academic abilities. The Rasch model from psychometrics is a popular generative probabilistic model that relates latent student skill, latent item difficulty, and observed student-item responses within a principled, unified framework. According to scholarly educational theory, discovered topics can be regarded as appropriate measurement items if (1) students' participation across the discovered topics is well fit by the Rasch model, and if (2) the topics are interpretable to subject-matter experts as being educationally meaningful. Such Rasch-scaled topics, with associated difficulty levels, could be of potential benefit to curriculum refinement, student assessment and personalised feedback. The technical challenge that remains, is to discover meaningful topics that simultaneously achieve good statistical fit with the Rasch model. To address this challenge, we combine the Rasch model with non-negative matrix factorisation based topic modelling, jointly fitting both models. We demonstrate the suitability of our approach with quantitative experiments on data from three Coursera MOOCs, and with qualitative survey results on topic interpretability on a Discrete Optimisation MOOC.

cs.LG↗

arXiv2017

End-to-End Differentially-Private Parameter Tuning in Spatial Histograms

Differentially-private histograms have emerged as a key tool for location privacy. While past mechanisms have included theoretical & experimental analysis, it has recently been observed that much of the existing literature does not fully provide differential privacy. The missing component, private parameter tuning, is necessary for rigorous evaluation of these mechanisms. Instead works frequently tune on training data to optimise parameters without consideration of privacy; in other cases selection is performed arbitrarily and independent of data, degrading utility. We address this open problem by deriving a principled tuning mechanism that privately optimises data-dependent error bounds. Theoretical results establish privacy and utility while extensive experimentation demonstrates that we can practically achieve true end-to-end privacy.

cs.DB↗

arXiv2016

The Bernstein Mechanism: Function Release under Differential Privacy

We address the problem of general function release under differential privacy, by developing a functional mechanism that applies under the weak assumptions of oracle access to target function evaluation and sensitivity. These conditions permit treatment of functions described explicitly or implicitly as algorithmic black boxes. We achieve this result by leveraging the iterated Bernstein operator for polynomial approximation of the target function, and polynomial coefficient perturbation. Under weak regularity conditions, we establish fast rates on utility measured by high-probability uniform approximation. We provide a lower bound on the utility achievable for any functional mechanism that is $\varepsilon$-differentially private. The generality of our mechanism is demonstrated by the analysis of a number of example learners, including naive Bayes, non-parametric estimators and regularized empirical risk minimization. Competitive rates are demonstrated for kernel density estimation; and $\varepsilon$-differential privacy is achieved for a broader class of support vector machines than known previously.

cs.DS↗

arXiv2016

Large-Scale Strategic Games and Adversarial Machine Learning

Decision making in modern large-scale and complex systems such as communication networks, smart electricity grids, and cyber-physical systems motivate novel game-theoretic approaches. This paper investigates big strategic (non-cooperative) games where a finite number of individual players each have a large number of continuous decision variables and input data points. Such high-dimensional decision spaces and big data sets lead to computational challenges, relating to efforts in non-linear optimization scaling up to large systems of variables. In addition to these computational challenges, real-world players often have limited information about their preference parameters due to the prohibitive cost of identifying them or due to operating in dynamic online settings. The challenge of limited information is exacerbated in high dimensions and big data sets. Motivated by both computational and information limitations that constrain the direct solution of big strategic games, our investigation centers around reductions using linear transformations such as random projection methods and their effect on Nash equilibrium solutions. Specific analytical results are presented for quadratic games and approximations. In addition, an adversarial learning game is presented where random projection and sampling schemes are investigated.

cs.GT↗

arXiv2015

MOOCs Meet Measurement Theory: A Topic-Modelling Approach

This paper adapts topic models to the psychometric testing of MOOC students based on their online forum postings. Measurement theory from education and psychology provides statistical models for quantifying a person's attainment of intangible attributes such as attitudes, abilities or intelligence. Such models infer latent skill levels by relating them to individuals' observed responses on a series of items such as quiz questions. The set of items can be used to measure a latent skill if individuals' responses on them conform to a Guttman scale. Such well-scaled items differentiate between individuals and inferred levels span the entire range from most basic to the advanced. In practice, education researchers manually devise items (quiz questions) while optimising well-scaled conformance. Due to the costly nature and expert requirements of this process, psychometric testing has found limited use in everyday teaching. We aim to develop usable measurement models for highly-instrumented MOOC delivery platforms, by using participation in automatically-extracted online forum topics as items. The challenge is to formalise the Guttman scale educational constraint and incorporate it into topic models. To favour topics that automatically conform to a Guttman scale, we introduce a novel regularisation into non-negative matrix factorisation-based topic modelling. We demonstrate the suitability of our approach with both quantitative experiments on three Coursera MOOCs, and with a qualitative survey of topic interpretability on two MOOCs by domain expert interviews.

cs.LG↗

arXiv2014

Principled Graph Matching Algorithms for Integrating Multiple Data Sources

This paper explores combinatorial optimization for problems of max-weight graph matching on multi-partite graphs, which arise in integrating multiple data sources. Entity resolution-the data integration problem of performing noisy joins on structured data-typically proceeds by first hashing each record into zero or more blocks, scoring pairs of records that are co-blocked for similarity, and then matching pairs of sufficient similarity. In the most common case of matching two sources, it is often desirable for the final matching to be one-to-one (a record may be matched with at most one other); members of the database and statistical record linkage communities accomplish such matchings in the final stage by weighted bipartite graph matching on similarity scores. Such matchings are intuitively appealing: they leverage a natural global property of many real-world entity stores-that of being nearly deduped-and are known to provide significant improvements to precision and recall. Unfortunately unlike the bipartite case, exact max-weight matching on multi-partite graphs is known to be NP-hard. Our two-fold algorithmic contributions approximate multi-partite max-weight matching: our first algorithm borrows optimization techniques common to Bayesian probabilistic inference; our second is a greedy approximation algorithm. In addition to a theoretical guarantee on the latter, we present comparisons on a real-world ER problem from Bing significantly larger than typically found in the literature, publication data, and on a series of synthetic problems. Our results quantify significant improvements due to exploiting multiple sources, which are made possible by global one-to-one constraints linking otherwise independent matching sub-problems. We also discover that our algorithms are complementary: one being much more robust under noise, and the other being simple to implement and very fast to run.

cs.DB↗

arXiv2014

Security Evaluation of Support Vector Machines in Adversarial Environments

Support Vector Machines (SVMs) are among the most popular classification techniques adopted in security applications like malware detection, intrusion detection, and spam filtering. However, if SVMs are to be incorporated in real-world security systems, they must be able to cope with attack patterns that can either mislead the learning algorithm (poisoning), evade detection (evasion), or gain information about their internal parameters (privacy breaches). The main contributions of this chapter are twofold. First, we introduce a formal general framework for the empirical evaluation of the security of machine-learning systems. Second, according to our framework, we demonstrate the feasibility of evasion, poisoning and privacy attacks against SVMs in real-world security problems. For each attack technique, we evaluate its impact and discuss whether (and how) it can be countered through an adversary-aware design of SVMs. Our experiments are easily reproducible thanks to open-source code that we have made available, together with all the employed datasets, on a public repository.

cs.LG↗

arXiv2014

Bounding Embeddings of VC Classes into Maximum Classes

One of the earliest conjectures in computational learning theory-the Sample Compression conjecture-asserts that concept classes (equivalently set systems) admit compression schemes of size linear in their VC dimension. To-date this statement is known to be true for maximum classes---those that possess maximum cardinality for their VC dimension. The most promising approach to positively resolving the conjecture is by embedding general VC classes into maximum classes without super-linear increase to their VC dimensions, as such embeddings would extend the known compression schemes to all VC classes. We show that maximum classes can be characterised by a local-connectivity property of the graph obtained by viewing the class as a cubical complex. This geometric characterisation of maximum VC classes is applied to prove a negative embedding result which demonstrates VC-d classes that cannot be embedded in any maximum class of VC dimension lower than 2d. On the other hand, we show that every VC-d class C embeds in a VC-(d+D) maximum class where D is the deficiency of C, i.e., the difference between the cardinalities of a maximum VC-d class and of C. For VC-2 classes in binary n-cubes for 4 <= n <= 6, we give best possible results on embedding into maximum classes. For some special classes of Boolean functions, relationships with maximum classes are investigated. Finally we give a general recursive procedure for embedding VC-d classes into VC-(d+k) maximum classes for smallest k.

cs.LG↗

arXiv2012

Scaling Multiple-Source Entity Resolution using Statistically Efficient Transfer Learning

We consider a serious, previously-unexplored challenge facing almost all approaches to scaling up entity resolution (ER) to multiple data sources: the prohibitive cost of labeling training data for supervised learning of similarity scores for each pair of sources. While there exists a rich literature describing almost all aspects of pairwise ER, this new challenge is arising now due to the unprecedented ability to acquire and store data from online sources, features driven by ER such as enriched search verticals, and the uniqueness of noisy and missing data characteristics for each source. We show on real-world and synthetic data that for state-of-the-art techniques, the reality of heterogeneous sources means that the number of labeled training data must scale quadratically in the number of sources, just to maintain constant precision/recall. We address this challenge with a brand new transfer learning algorithm which requires far less training data (or equivalently, achieves superior accuracy with the same data) and is trained using fast convex optimization. The intuition behind our approach is to adaptively share structure learned about one scoring problem with all other scoring problems sharing a data source in common. We demonstrate that our theoretically motivated approach incurs no runtime cost while it can maintain constant precision/recall with the cost of labeling increasing only linearly with the number of sources.

cs.DB↗

arXiv2012

A Bayesian Approach to Discovering Truth from Conflicting Sources for Data Integration

In practical data integration systems, it is common for the data sources being integrated to provide conflicting information about the same entity. Consequently, a major challenge for data integration is to derive the most complete and accurate integrated records from diverse and sometimes conflicting sources. We term this challenge the truth finding problem. We observe that some sources are generally more reliable than others, and therefore a good model of source quality is the key to solving the truth finding problem. In this work, we propose a probabilistic graphical model that can automatically infer true records and source quality without any supervision. In contrast to previous methods, our principled approach leverages a generative process of two types of errors (false positive and false negative) by modeling two different aspects of source quality. In so doing, ours is also the first approach designed to merge multi-valued attribute types. Our method is scalable, due to an efficient sampling-based inference algorithm that needs very few iterations in practice and enjoys linear time complexity, with an even faster incremental variant. Experiments on two real world datasets show that our new method outperforms existing state-of-the-art approaches to the truth finding problem.

cs.DB↗