Biblio

According to the theory of contextual integrity (CI), privacy norms prescribeinformation flows with reference to five parameters — sender, recipient, subject, information type, and transmission principle. Because privacy is grasped contextually (e.g., health, education, civic life, etc.), the values of these parameters range over contextually meaningful ontologies — of information types (or topics) and actors (subjects, senders, and recipients), incontextually defined capacities. As an alternative to predominant approaches to privacy, which were ineffective against novel information practices enabled by IT, CI was able both to pinpoint sources of disruption andprovide grounds for either accepting or rejecting them. Mounting challengesfrom a burgeoning array of networked, sensor-enabled devices (IoT) and data-ravenous machine learning systems, similar in form though magnified in scope, call for renewed attention to theory. This Article introduces themetaphor of a data (food) chain to capture the nature of these challenges.With motion up the chain, where data of higher order is inferred from lower-order data, the crucial question is whether privacy norms governing lower-order data are sufficient for the inferred higher-order data. While CI has a response to this question, a greater challenge comes from data primitives, such as digital impulses of mouse clicks, motion detectors, and bare GPS coordinates, because they appear to have no meaning. Absent a semantics, they escape CI’s privacy norms entirely.

Modern enterprises rely on Data Leakage Prevention (DLP) systems to enforce privacy policies that prevent unintentional flow of sensitive information to unauthorized entities. However, these systems operate based on rule sets that are limited to syntactic analysis and therefore completely ignore the semantic relationships between participants involved in the information exchanges. For similar reasons, these systems cannot enforce complex privacy policies that require temporal reasoning about events that have previously occurred. 

To address these limitations, we advocate a new design methodology for DLP systems centered on the notion of Contextual Integrity (CI). We use the CI framework to abstract real-world communication exchanges into formally defined information flows where privacy policies describe sequences of admissible flows. CI allows us to decouple (1) the syntactic extraction of flows from information exchanges, and (2) the enforcement of privacy policies on these flows. We applied this approach to built VACCINE, a DLP auditing system for emails. VACCINE uses state-of-the-art techniques in natural language processing to extract flows from email text. It also provides a declarative language for describing privacy policies. These policies are automatically compiled to operational rules that the system uses for detecting data leakages. We evaluated VACCINE on the Enron email corpus and show that it improves over the state of the art both in terms of the expressivity of the policies that DLP systems can enforce as well as its precision in detecting data leakages.

Images perturbed subtly to be misclassified by neural networks, called adversarial examples, have emerged as a technically deep challenge and an important concern for several application domains. Most research on adversarial examples takes as its only constraint that the perturbed images are similar to the originals. However, real-world application of these ideas often requires the examples to satisfy additional objectives, which are typically enforced through custom modifications of the perturbation process. In this article, we propose adversarial generative nets (AGNs), a general methodology to train a generator neural network to emit adversarial examples satisfying desired objectives. We demonstrate the ability of AGNs to accommodate a wide range of objectives, including imprecise ones difficult to model, in two application domains. In particular, we demonstrate physical adversarial examples—eyeglass frames designed to fool face recognition—with better robustness, inconspicuousness, and scalability than previous approaches, as well as a new attack to fool a handwritten-digit classifier.

As Cyber-Physical Systems (CPSs) become more autonomous, it becomes harder for humans who interact with the CPSs to understand the behavior of the systems. Particularly for CPSs that must perform tasks while optimizing for multiple quality objectives and acting under uncertainty, it can be difficult for humans to understand the system behavior generated by an automated planner. This work-in-progress presents an approach at clarifying system behavior through interactive explanation by allowing end-users to ask Why and Why-Not questions about specific behaviors of the system, and providing answers in the form of contrastive explanation.

We present Copland, a language for specifying layered attestations. Layered attestations provide a remote appraiser with structured evidence of the integrity of a target system to support a trust decision. The language is designed to bridge the gap between formal analysis of attestation security guarantees and concrete implementations. We therefore provide two semantic interpretations of terms in our language. The first is a denotational semantics in terms of partially ordered sets of events. This directly connects Copland to prior work on layered attestation. The second is an operational semantics detailing how the data and control flow are executed. This gives explicit implementation guidance for attestation frameworks. We show a formal connection between the two semantics ensuring that any execution according to the operational semantics is consistent with the denotational event semantics. This ensures that formal guarantees resulting from analyzing the event semantics will hold for executions respecting the operational semantics. All results have been formally verified with the Coq proof assistant.

Many cybersecurity breaches occur due to users not following security regulations, chief among them regulations pertaining to what might be termed hygiene---including applying software patches to operating systems, updating software applications, and maintaining strong passwords. 

We capture cybersecurity expectations on users as norms. We empirically investigate sanctioning mechanisms in promoting compliance with those norms as well as the detrimental effect of sanctions on the ability of users to complete their work. We do so by developing a game that emulates the decision making of workers in a research lab. 

We find that relative to group sanctions, individual sanctions are more effective in achieving compliance and less detrimental on the ability of users to complete their work.
Our findings have implications for workforce training in cybersecurity.

This chapter explores the relationship between the concept of emergence, the goal of theoretical completeness, and the Principle of Sufficient Reason. Samuel Alexander and C. D. Broad argued for limits to the power of scientific explanation. Chemical explanation played a central role in their thinking. After Schrödinger’s work in the 1920s their examples seem to fall flat. However, there are more general lessons from the emergentists that need to be explored. There are cases where we know that explanation of some phenomenon is impossible. What are the implications of known limits to the explanatory power of science, and the apparent ineliminability of brute facts for emergence? One lesson drawn here is that we must embrace a methodological rather than a metaphysical conception of the Principle of Sufficient Reason.

Blockchains are one solution for secure distributed interaction, but security vulnerabilities have already been exposed in existing programs. Obsidian, a new blockchain programming language, seeks to prevent some of these vulnerabilities using typestate and linearity. We evaluate the current design of Obsidian by implementing a blockchain application for parametric insurance as a case study. We compare this implementation to one written in Solidity, and find that Obsidian can provide stronger safety guarantees.

Approaches for programming language design used commonly in the research community today center around theoretical and performance-oriented evaluation. Recently, researchers have been considering more approaches to language design, including the use of quantitative and qualitative user studies that examine how different designs might affect programmers. In this paper, we argue for an interdisciplinary approach that incorporates many different methods in the creation and evaluation of programming languages. We argue that the addition of user-oriented design techniques can be helpful at many different stages in the programming language design process.

Due to the evolution of programming languages, interpreted languages have gained widespread use in scientific and research computing. Interpreted languages excel at being portable, easy to use, and fast in prototyping than their ahead-of-time (AOT) counterparts, including C, C++, and Fortran. While traditionally considered as slow to execute, advancements in Just-in-Time (JIT) compilation techniques have significantly improved the execution speed of interpreted languages and in some cases outperformed AOT languages. In this paper, we explore some challenges and design strategies in developing a high performance parallel discrete event simulation engine, called Simian, written with interpreted languages with JIT capabilities, including Python, Lua, and Javascript. Our results show that Simian with JIT performs similarly to AOT simulators, such as MiniSSF and ROSS. We expect that with features like good performance, userfriendliness, and portability, the just-in-time parallel simulation will become a common choice for modeling and simulation in the near future.
 

Adaptive systems are expected to adapt to unanticipated run-time events using imperfect information about themselves, their environment, and goals. This entails handling the effects of uncertainties in decision-making, which are not always considered as a first-class concern. This paper contributes a formal analysis technique that explicitly considers uncertainty in sensing when reasoning about the best way to adapt, together with uncertainty reduction mechanisms to improve system utility. We illustrate our approach on a Denial of Service (DoS) attack scenario and present results that demonstrate the benefits of uncertainty-aware decision-making in comparison to using an uncertainty-ignorant approach, both in the presence and absence of uncertainty reduction mechanisms.

Software-defined networking (SDN) enables efficient networkmanagement. As the technology matures, utilities are looking to integrate those benefits to their operations technology (OT) networks. To help the community to better understand and evaluate the effects of such integration, we develop DSSnet, a testing platform that combines a power distribution system simulator and an SDN-based network emulator for smart grid planning and evaluation. DSSnet relies on a container-based virtual time system to achieve efficient synchronization between the simulation and emulation systems. To enhance the system scalability and usability, we extend DSSnet to support a distributed controller environment. To enhance system fidelity, we extend the virtual time system to support kernel-based switches. We also evaluate the system performance of DSSnet and demonstrate the usability of DSSnet with a resilient demand response application case study.

Network operators face a challenge of ensuring correctness as networks grow more complex, in terms of scale and increasingly in terms of diversity of software components. Network-wide verification approaches can spot errors, but assume a simplified abstraction of the functionality of individual network devices, which may deviate from the real implementation. In this paper, we propose a technique for high-coverage testing of end-to-end network correctness using the real software that is deployed in these networks. Our design is effectively a hybrid, using an explicit-state model checker to explore all network-wide execution paths and event orderings, but executing real software as subroutines for each device. We show that this approach can detect correctness issues that would be missed both by existing verification and testing approaches, and a prototype implementation suggests the technique can scale to larger networks
with reasonable performance.

Abstract—Network intrusion detection systems (NIDS) are essential security building-blocks for today’s organizations to ensure safe and trusted communication of information. In this paper, we study the feasibility of off-line deep learning based NIDSes by constructing the detection engine with multiple advanced deep learning models and conducting a quantitative and comparative evaluation of those models. We first introduce the general deep learning methodology and its potential implication on the network intrusion detection problem. We then review multiple machine learning solutions to two network intrusion detection tasks (NSL-KDD and UNSW-NB15 datasets). We develop a TensorFlow-based deep learning library, called NetLearner, and implement a handful of cutting-edge deep learning models for NIDS. Finally, we conduct a quantitative and comparative performance evaluation of those models using NetLearner.

Software-defined networking (SDN) continues to grow in popularity because of its programmable and extensible control plane realized through network applications (apps). However, apps introduce significant security challenges that can systemically disrupt network operations, since apps must access or modify data in a shared control plane state. If our understanding of how such data propagate within the control plane is inadequate, apps can co-opt other apps, causing them to poison the control plane’s integrity. 

We present a class of SDN control plane integrity attacks that we call cross-app poisoning (CAP), in which an unprivileged app manipulates the shared control plane state to trick a privileged app into taking actions on its behalf. We demonstrate how role-based access control (RBAC) schemes are insufficient for preventing such attacks because they neither track information flow nor enforce information flow control (IFC). We also present a defense, ProvSDN, that uses data provenance to track information flow and serves as an online reference monitor to prevent CAP attacks. We implement ProvSDN on the ONOS SDN controller and demonstrate that information flow can be tracked with low-latency overheads.

We present DeepPicar, a low-cost deep neural network based autonomous car platform. DeepPicar is a small scale replication of a real self-driving car called DAVE-2 by NVIDIA. DAVE-2 uses a deep convolutional neural network (CNN), which takes images from a front-facing camera as input and produces car steering angles as output. DeepPicar uses the same network architecture—9 layers, 27 million connections and 250K parameters—and can drive itself in real-time using a web camera and a Raspberry Pi 3 quad-core platform. Using DeepPicar, we analyze the Pi 3’s computing capabilities to support end-to-end deep learning based real-time control of autonomous vehicles. We also systematically compare other contemporary embedded computing platforms using the DeepPicar’s CNN-based real-time control workload. We find that all tested platforms, including the Pi 3, are capable of supporting the CNN-based real-time control, from 20 Hz up to 100 Hz, depending on hardware platform. However, we find that shared resource contention remains an important issue that must be considered in applying CNN models on shared memory based embedded computing platforms; we observe up to 11.6X execution time increase in the CNN based control loop due to shared resource contention. To protect the CNN workload, we also evaluate state-of-the-art cache partitioning and memory bandwidth throttling techniques on the Pi 3. We find that cache partitioning is ineffective, while memory bandwidth throttling is an effective solution.

As the Industrial Internet of Things (IIot) becomes more prevalent in critical application domains, ensuring security and resilience in the face of cyber-attacks is becoming an issue of paramount importance. Cyber-attacks against critical infrastructures, for example, against smart water-distribution and transportation systems, pose serious threats to public health and safety. Owing to the severity of these threats, a variety of security techniques are available. However, no single technique can address the whole spectrum of cyber-attacks that may be launched by a determined and resourceful attacker. In light of this, we consider a multi-pronged approach for designing secure and resilient IIoT systems, which integrates redundancy, diversity, and hardening techniques. We introduce a framework for quantifying cyber-security risks and optimizing IIoT design by determining security investments in redundancy, diversity, and hardening. To demonstrate the applicability of our framework, we present two case studies in water distribution and transportation a case study in water-distribution systems. Our numerical evaluation shows that integrating redundancy, diversity, and hardening can lead to reduced security risk at the same cost.

Attacks in cyber-physical systems (CPS) which manipulate sensor readings can cause enormous physical damage if undetected. Detection of attacks on sensors is crucial to mitigate this issue. We study supervised regression as a means to detect anomalous sensor readings, where each sensor's measurement is predicted as a function of other sensors. We show that several common learning approaches in this context are still vulnerable to \emph{stealthy attacks}, which carefully modify readings of compromised sensors to cause desired damage while remaining undetected. Next, we model the interaction between the CPS defender and attacker as a Stackelberg game in which the defender chooses detection thresholds, while the attacker deploys a stealthy attack in response. We present a heuristic algorithm for finding an approximately optimal threshold for the defender in this game, and show that it increases system resilience to attacks without significantly increasing the false alarm rate.

In the past couple of years, railway infrastructure has been growing more connected, resembling more of a traditional Cyber-Physical System model. Due to the tightly coupled nature between the cyber and physical domains, new attack vectors are emerging that create an avenue for remote hijacking of system components not designed to withstand such attacks. As such, best practice cybersecurity techniques need to be put in place to ensure the safety and resiliency of future railway designs, as well as infrastructure already in the field. However, traditional large-scale experimental evaluation that involves evaluating a large set of variables by running a design of experiments (DOE) may not always be practical and might not provide conclusive results. In addition, to achieve scalable experimentation, the modeling abstractions, simulation configurations, and experiment scenarios must be designed according to the analysis goals of the evaluations. Thus, it is useful to target a set of key operational metrics for evaluation and configure and extend the traditional DOE methods using these metrics. In this work, we present a metrics-driven evaluation approach for evaluating the security and resilience of railway critical infrastructure using a distributed simulation framework. A case study with experiment results is provided that demonstrates the capabilities of our testbed.

The explosive proliferation of Internet of Things (IoT) devices is generating an incomprehensible amount of data. Machine learning plays an imperative role in aggregating this data and extracting valuable information for improving operational and decision-making processes. In particular, emerging machine intelligence platforms that host pre-trained machine learning models are opening up new opportunities for IoT industries. While those platforms facilitate customers to analyze IoT data and deliver faster and accurate insights, end users and machine learning service providers (MLSPs) have raised concerns regarding security and privacy of IoT data as well as the pre-trained machine learning models for certain applications such as healthcare, smart energy, etc. In this paper, we propose a cloud-assisted, privacy-preserving machine learning classification scheme over encrypted data for IoT devices. Our scheme is based on a three-party model coupled with a two-stage decryption Paillier-based cryptosystem, which allows a cloud server to interact with MLSPs on behalf of the resource-constrained IoT devices in a privacy-preserving manner, and shift load of computation-intensive classification operations from them. The detailed security analysis and the extensive simulations with different key lengths and number of features and classes demonstrate that our scheme can effectively reduce the overhead for IoT devices in machine learning classification applications.

Stealthy attackers often disable or tamper with system monitors to hide their tracks and evade detection. In this poster, we present a data-driven technique to detect such monitor compromise using evidential reasoning. Leveraging the fact that hiding from multiple, redundant monitors is difficult for an attacker, to identify potential monitor compromise, we combine alerts from different sets of monitors by using Dempster-Shafer theory, and compare the results to find outliers. We describe our ongoing work in this area.

In this paper, we analyze the security of cyber-physical systems using the ADversary VIew Security Evaluation (ADVISE) meta modeling approach, taking into consideration the efects of physical attacks. To build our model of the system, we construct an ontology that describes the system components and the relationships among them. The ontology also deines attack steps that represent cyber and physical actions that afect the system entities. We apply the ADVISE meta modeling approach, which admits as input our deined ontology, to a railway system use case to obtain insights regarding the system’s security. The ADVISE Meta tool takes in a system model of a railway station and generates an attack execution graph that shows the actions that adversaries may take to reach their goal. We consider several adversary proiles, ranging from outsiders to insider staf members, and compare their attack paths in terms of targeted assets, time to achieve the goal, and probability of detection. The generated results show that even adversaries with access to noncritical assets can afect system service by intelligently crafting their attacks to trigger a physical sequence of efects. We also identify the physical devices and user actions that require more in-depth monitoring to reinforce the system’s security.

Many of the emerging wide-area monitoring protection and control (WAMPAC) applications in modern electrical grids rely heavily on the availability and integrity of widespread phasor measurement unit (PMU) data. Therefore, it is critical to protect PMU networks against growing cyber-attacks and system faults. In this paper, we present a self-healing PMU network design that considers both power system observability and communication network characteristics. Our design utilizes centralized network control, such as the emerging software-defined networking (SDN) technology, to design resilient network self-healing algorithms against cyber-attacks. Upon detection of a cyber-attack, the PMU network can reconfigure itself to isolate compromised devices and re-route measurement
data with the goal of preserving the power system observability. We have developed a proof-of-concept system in a container-based network testbed using integer linear programming to solve a graphbased PMU system model.We also evaluate the system performance regarding the self-healing plan generation and installation using the IEEE 30-bus system.