Helmholtz Center for Information Security
CISPA – Helmholtz-Zentrum für InformationssicherheitNot a member yet
3406 research outputs found
Sort by
Membership Inference Attacks Against Recommender Systems
Get PDFRecently, recommender systems have achieved promising performances and become one of the most widely used web applications. However, recommender systems are often trained on highly sensitive user data, thus potential data leakage from recommender systems may lead to severe privacy problems.
In this paper, we make the first attempt on quantifying the privacy leakage of recommender systems through the lens of membership inference. In contrast with traditional membership inference against machine learning classifiers, our attack faces two main differences. First, our attack is on the user-level but not on the data sample-level. Second, the adversary can only observe the ordered recommended items from a recommender system instead of prediction results in the form of posterior probabilities. To address the above challenges, we propose a novel method by representing users from relevant items. Moreover, a shadow recommender is established to derive the labeled training data for training the attack model. Extensive experimental results show that our attack framework achieves a strong performance. In addition, we design a defense mechanism to effectively mitigate the membership inference threat of recommender systems
Exorcising Spectres with secure compilers
Get PDFAttackers can access sensitive information of programs by exploiting the side-effects of speculatively-executed instructions using Spectre attacks. To mitigate these attacks, popular compilers deployed a wide range of countermeasures whose security, however, has not been ascertained: while some are believed to be secure, others are known to be insecure and result in vulnerable programs.
This paper develops formal foundations for reasoning about the
security of these defenses. For this, it proposes a framework of secure compilation criteria that characterise when compilers produce code resistant against Spectre v1 attacks. With this framework, this paper performs a comprehensive security analysis of countermeasures against Spectre v1 attacks implemented in major compilers, deriving the first security proofs of said countermeasures
Delta-based Verification of Software Product Families
Get PDFThe quest for feature- and family-oriented deductive verification of software product lines resulted in several proposals. In this paper we look at delta-oriented modeling of product lines and combine two new ideas: first, we extend Hähnle & Schaefer’s delta-oriented version of Liskov’s substitution principle for behavioral subtyping to work also for overridden behavior in benign cases. For this to succeed, programs need to be in a certain normal form. The required normal form turns out to be achievable in many cases by a set of program transformations, whose correctness is ensured by the recent technique of abstract execution. This is a generalization of symbolic execution that permits reasoning about abstract code elements. It is needed, because code deltas contain partially unknown code contexts in terms of “original” calls. Second, we devise a modular verification procedure for deltas based on abstract execution, representing deltas as abstract programs calling into unknown contexts. The result is a “delta-based” verification approach, where each modification of a method in a code delta is verified in isolation, but which overcomes the strict limitations of behavioral subtyping and works for many practical programs. The latter claim is substantiated with case studies and benchmarks
Robust Monitoring for Medical Cyber-Physical Systems
Get PDFSome medical implants act autonomously: they assess the current health status of a patient and administer treatment when appropriate. An improper treatment, however, can cause serious harm. Here, the decision logic leading to the treatment relies on data obtained from sensors — an inherently imperfect medium. Cop- ing with these inaccuracies requires the logic to be robust in the sense that slight perturbations in the measurements do not significantly alter the decision. Determining the extent to which an algorithm is robust automatically does not scale well for complex and opaque components. This is particularly problematic when ma- chine learning is involved. Yet, the analysis is feasible for simpler safety-related components such as a runtime monitor, which ob- serves the system and intervenes in a treatment when necessary. Its significantly lower complexity generally allows for providing static guarantees on the runtime behavior of the monitor. Complementing these guarantees with a robustness analysis constitutes a major step toward certifiable medical cyber-physical systems con- trolled by opaque, machine-learned components. Hence, this paper reports on ongoing research in the direction of a robustness analysis for the runtime monitoring framework RTLola
Tardigrade: An Atomic Broadcast Protocol for Arbitrary Network Conditions
Get PDFWe study the problem of \emph{atomic broadcast}---the underlying problem addressed by blockchain protocols---in the presence of a malicious adversary who corrupts some fraction of the parties running the protocol.
Existing protocols are either robust for any number of corruptions in a
\emph{synchronous} network (where
messages are delivered within some known
time~) but fail if the synchrony assumption is violated, or tolerate fewer than corrupted parties in an
\emph{asynchronous} network (where messages can be delayed arbitrarily) and cannot tolerate more corruptions even if the network happens to be well behaved.
%Although protocols for the latter case give seemingly stronger guarantees, this is not the case since they (inherently) tolerate a lower fraction of corrupted parties.
\smallskip
We design an atomic broadcast protocol (\name) that, for any with , provides security against corrupted parties if the network is synchronous, while remaining secure when parties are corrupted even in an asynchronous network.
We show that \name~achieves optimal tradeoffs between and~. Finally, we show
a second protocol (\nametwo) with similar (but slightly weaker) guarantees that achieves per-transaction
communication complexity linear in~
On the Computational Tractability of a Geographic Clustering Problem Arising in Redistricting
Get PDFRedistricting is the problem of dividing up a state into a given number k of regions (called districts) where the voters in each district are to elect a representative. The three primary criteria are: that each district be connected, that the populations of the districts be equal (or nearly equal), and that the districts are "compact". There are multiple competing definitions of compactness, usually minimizing some quantity.
One measure that has been recently been used is number of cut edges. In this formulation of redistricting, one is given atomic regions out of which each district must be built (e.g., in the U.S., census blocks). The populations of the atomic regions are given. Consider the graph with one vertex per atomic region and an edge between atomic regions with a shared boundary of positive length. Define the weight of a vertex to be the population of the corresponding region. A districting plan is a partition of vertices into k pieces so that the parts have nearly equal weights and each part is connected. The districts are considered compact to the extent that the plan minimizes the number of edges crossing between different parts.
There are two natural computational problems: find the most compact districting plan, and sample districting plans (possibly under a compactness constraint) uniformly at random.
Both problems are NP-hard so we consider restricting the input graph to have branchwidth at most w. (A planar graph’s branchwidth is bounded, for example, by its diameter.) If both k and w are bounded by constants, the problems are solvable in polynomial time. In this paper, we give lower and upper bounds that characterize the complexity of these problems in terms of parameters k and w. For simplicity of notation, assume that each vertex has unit weight. We would ideally like algorithms whose running times are of the form O(f(k,w) n^c) for some constant c independent of k and w (in which case the problems are said to be fixed-parameter tractable with respect to those parameters). We show that, under standard complexity-theoretic assumptions, no such algorithms exist. However, the problems are fixed-parameter tractable with respect to each of these parameters individually: there exist algorithms with running times of the form O(f(k) n^{O(w)}) and O(f(w) n^{k+1}). The first result was previously known. The new one, however, is more relevant to the application to redistricting, at least for coarse instances. Indeed, we have implemented a version of the algorithm and have used to successfully find optimally compact solutions to all redistricting instances for France (except Paris, which operates under different rules) under various population-balance constraints. For these instances, the values for w are modest and the values for k are very small
Adversarial Watermarking Transformer: Towards Tracing Text Provenance with Data Hiding
No full textRecent advances in natural language generation have introduced powerful language models with high-quality output text. However, this raises concerns about the potential misuse of such models for malicious purposes. In this paper, we study natural language watermarking as a defense to help better mark and trace the provenance of text. We introduce the Adversarial Watermarking Transformer (AWT) with a jointly trained encoder-decoder and adversarial training that, given an input text and a binary message, generates an output text that is unobtrusively encoded with the given message. We further study different training and inference strategies to achieve minimal changes to the semantics and correctness of the input text.
AWT is the first end-to-end model to hide data in text by automatically learning -- without ground truth -- word substitutions along with their locations in order to encode the message. We empirically show that our model is effective in largely preserving text utility and decoding the watermark while hiding its presence against adversaries. Additionally, we demonstrate that our method is robust against a range of attacks
Why Eve and Mallory Still Love Android: Revisiting TLS (In)Security in Android Applications
Get PDFAndroid applications have a long history of being vulnerable to man-in-the-middle attacks due to insecure custom TLS certificate validation implementations. To resolve this, Google deployed the Network Security Configuration (NSC), a configuration-based approach to increase custom certificate validation logic security, and implemented safeguards in Google Play to block insecure applications. In this paper, we perform a large-scale in-depth investigation of the effectiveness of these countermeasures: First, we investigate the security of 99,212 NSC settings files in 1,335,322 Google Play apps using static code and manual analysis techniques. We find that 88.87% of the apps using custom NSC settings downgrade security compared to the default settings, and only 0.67% implement certificate pinning. Second, we penetrate Google Play’s protection mechanisms by trying to publish apps that are vulnerable to man-in-the-middle attacks. In contrast to official announcements by Google, we found that Play does not effectively block vulnerable apps. Finally, we performed a static code analysis study of 15,000 apps and find that 5,511 recently published apps still contain vulnerable certificate validation code. Overall, we attribute most of the problems we find to insufficient support for developers, missing clarification of security risks in official documentation, and inadequate security checks for vulnerable applications in Google Play
Exploring User-Centered Security Design for Usable Authentication Ceremonies
Get PDFSecurity technology often follows a systems design approach that focuses on components instead of users. As a result, the users' needs and values are not sufficiently addressed, which has implications on security usability. In this paper, we report our lessons learned from applying a user-centered security design process to a well-understood security usability challenge, namely key authentication in secure instant messaging. Users rarely perform these key authentication ceremonies, which makes their end-to-end encrypted communication vulnerable. Our approach includes collaborative design workshops, an expert evaluation, iterative storyboard prototyping, and an online evaluation.
While we could not demonstrate that our design approach resulted in improved usability or user experience, we found that user-centered prototypes can increase the users' comprehension of security implications. Hence, prototypes based on users' intuitions, needs, and values are useful starting points for approaching long-standing security challenges. Applying complementary design approaches may improve usability and user experience further
LIGHTBLUE: Automatic Profile-Aware Debloating of Bluetooth Stacks
Get PDFThe Bluetooth standard is ubiquitously supported by com-
puters, smartphones, and IoT devices. Due to its complex-
ity, implementations require large codebases, that are prone
to security vulnerabilities, such as the recently discovered
BlueBorne and BadBluetooth attacks. While defined by the
standard, most of the Bluetooth functionality, as defined by
different Bluetooth profiles, is not required in the common
usage scenarios.
Starting from this observation, we implement L IGHT B LUE ,
a framework performing automatic, profile-aware debloating
of Bluetooth stacks, allowing users to automatically minimize
their Bluetooth attack surface by removing unneeded Blue-
tooth features. L IGHT B LUE starts with a target Bluetooth
application, detects the associated Bluetooth profiles, and ap-
plies a combination of control-flow and data-flow analysis
to remove unused code within a Bluetooth host code. Fur-
thermore, to debloat the Bluetooth firmware, L IGHT B LUE
extracts the used Host Controller Interface (HCI) commands
and patches the HCI dispatcher in the Bluetooth firmware au-
tomatically, so that the Bluetooth firmware avoids processing
unneeded HCI commands.
We evaluate L IGHT B LUE on four different Bluetooth hosts
and three different Bluetooth controllers. Our evaluation
shows that L IGHT B LUE achieves between 32% and 50% code
reduction in the Bluetooth host code and between 57% and
83% HCI command reduction in the Bluetooth firmware. This
code reduction leads to the prevention of attacks responsible
for at least 20 CVEs, such as BlueBorne and BadBluetooth,
while introducing no performance overhead and without af-
fecting the behavior of the debloated application