Search

Derivatives occur frequently in computer graphics and arise in many different contexts. Gradients and often Hessians of objective functions are required for efficient optimization. Gradients of potential energy are used to compute forces. Constitutive models are frequently formulated from an energy density, which must be differentiated to compute stress. Hessians of potential energy or energy density are needed for implicit integration. As the methods used in computer graphics become more accurate and sophisticated, the complexity of the functions that must be differentiated also increases. The purpose of this course is to show that it is practical to compute derivatives even for functions that may seem impossibly complex. This course provides practical strategies and techniques for planning, computing, testing, debugging, and optimizing routines for computing first and second derivatives of real-world routines. This course will also introduce and explore auto differentiation, which encompasses a variety of techniques for obtaining derivatives automatically. The goal of this course is not to introduce the concept of derivatives, how to use them, or even how to calculate them per se. This is not intended to be a calculus course; we will assume that our audience is familiar with multivariable calculus. Instead, the emphasis is on implementing derivatives of complicated computational procedures in computer programs and actually getting them to work.

Online social networks (OSNs) have become a commodity in our daily life. As an important concept in sociology and viral marketing, the study of social influence has received a lot of attentions in academia. Most of the existing proposals work well on dominant OSNs, such as Twitter, since these sites are mature and many users have generated a large amount of data for the calculation of social influence. Unfortunately, cold-start users on emerging OSNs generate much less activity data, which makes it challenging to identify potential influential users among them. In this work, we propose a practical solution to predict whether a cold-start user will become an influential user on an emerging OSN, by opportunistically leveraging the user’s information on dominant OSNs. A supervised machine learning-based approach is adopted, transferring the knowledge of both the descriptive information and dynamic activities on dominant OSNs. Descriptive features are extracted from the public data on a user’s homepage. In particular, to extract useful information from the fine-grained dynamic activities that cannot be represented by the statistical indices, we use deep learning technologies to deal with the sequential activity data. Using the real data of millions of users collected from Twitter (a dominant OSN) and Medium (an emerging OSN), we evaluate the performance of our proposed framework to predict prospective influential users. Our system achieves a high prediction performance based on different social influence definitions.

Understanding complex information behaviour in an empirical, ecologically valid way is difficult at the best of times. It verges on impossible with artificial environments and tightly constrained tasks such as those found in usability labs. Similarly, artificial measures of user satisfaction that are readily available from test collections are unlikely to provoke revolutionary insights. The alternative is to study user behaviour in the wild, however this involves complex trade-offs between method, precision, privacy, interpretation and generalizability. This tutorial addresses both qualitative and quantitative methods, and how to balance the risks of studying unpredictable but fully human behaviour.

Allowing projections on moving objects is associated with a problem that a projection might shift due to the delay between tracking and projection. In the present paper, we proposed a new prediction model based on deep neural networks that can be used to predict both pose and position of the target object. As a result, we developed a real-time tracking and projection system named"MirAIProjection that employs motion-capture cameras and common projectors. We conducted several experiments to evaluate the effectiveness of the proposed system and demonstrated that the proposed system could reduce the slipping and increase the accuracy and robustness of the projection.

The gap between research and design practice has long been a concern for the HCI community. In this article, we explore how different translations of HCI knowledge might bridge this gap. A literature review characterizes the gap as having two key dimensions—one between general theory and particular artefacts and a second between academic HCI research and professional UX design practice. We report on a 5-year engagement between HCI researchers and a major media company to explore how a particular piece of HCI research, the trajectories conceptual framework , might be translated for and with UX practitioners. We present various translations of this framework and fit them into the gap we previously identified. This leads us to refine the idea of translations , suggesting that they may be led by researchers, by practitioners or co-produced by both as boundary objects . We consider the benefits of each approach.

Several proof assistants, such as Isabelle or Coq, can concurrently check multiple proofs. In contrast, the vast majority of today's small proof checkers either does not support concurrency at all or only limited forms thereof, restricting the efficiency of proof checking on multi-core processors. This work shows the design of a small, memory- and thread-safe kernel that efficiently checks proofs both concurrently and non-concurrently. This design is implemented in a new proof checker called Kontroli for the lambda-Pi calculus modulo rewriting, which is an established framework to uniformly express a multitude of logical systems. Kontroli is faster than the reference proof checker for this calculus, Dedukti, on all of five evaluated datasets obtained from proof assistants and interactive theorem provers. Furthermore, Kontroli reduces the time of the most time-consuming part of proof checking using eight threads by up to 6.6x., 11th ACM SIGPLAN International Conference on Certified Programs and Proofs (CPP '22), Jan 2022, Philadelphia, PA, United States

RAN virtualization will become a key technology for the last mile of next-generation mobile networks driven by initiatives such as the O-RAN alliance. However, due to the computing fluctuations inherent to wireless dynamics and resource contention in shared computing infrastructure, the price to migrate from dedicated to shared platforms may be too high. Indeed, we show in this paper that the baseline architecture of a base station��s distributed unit (DU) collapses upon moments of deficit in computing capacity. Recent solutions to accelerate some signal processing tasks certainly help but do not tackle the core problem: a DU pipeline that requires predictable computing to provide carrier-grade reliability. We present Nuberu, a novel pipeline architecture for 4G/5G DUs specifically engineered for non-deterministic computing platforms. Our design has one key objective to attain reliability: to guarantee a minimum set of signals that preserve synchronization between the DU and its users during computing capacity shortages and, provided this, maximize network throughput. To this end, we use techniques such as tight deadline control, jitter-absorbing buffers, predictive HARQ, and congestion control. Using an experimental prototype, we show that Nuberu attains >95% of the theoretical spectrum efficiency in hostile environments, where state-of-art approaches lose connectivity, and at least 80% resource savings, This work has been supported by the European Commission through Grant No. 101017109 (DAEMON project) and Grant No. 101015956 (Hexa-X project), and the CERCA Programme/Generalitat de Catalunya.

While they have long been a subject of academic study, social virtual reality (SVR) systems are now attracting increasingly large audiences on current consumer virtual reality systems. The design space of SVR systems is very large, and relatively little is known about how these systems should be constructed in order to be usable and efficient. In this paper we present Ubiq, a toolkit that focuses on facilitating the construction of SVR systems. We argue for the design strategy of Ubiq and its scope. Ubiq is built on the Unity platform. It provides core functionality of many SVR systems such as connection management, voice, avatars, etc. However, its design remains easy to extend. We demonstrate examples built on Ubiq and how it has been successfully used in classroom teaching. Ubiq is open source (Apache License) and thus enables several use cases that commercial systems cannot.

International audience; The control of an avatar’s facial expressions in virtual reality is mainly based on the automated recognition and transposition of the user’s facial expressions. These isomorphic techniques are limited to what users can convey with their own face and have recognition issues. To overcome these limitations, non-isomorphic techniques rely on interaction techniques using input devices to control the avatar’s facial expressions. Such techniques need to be designed to quickly and easily select and control an expression, and not disrupt a main task such as talking. We present the design of a set of new non- isomorphic interaction techniques for controlling an avatar facial expression in VR using a standard VR controller. These techniques have been evaluated through two controlled experiments to help designing an interaction technique combining the strengths of each approach. This technique was evaluated in a final ecological study showing it can be used in contexts such as social applications.

Federated Learning (FL) allows parties to learn a shared prediction model by delegating the training computation to clients and aggregating all the separately trained models on the server. To prevent private information being inferred from local models, Secure Aggregation (SA) protocols are used to ensure that the server is unable to inspect individual trained models as it aggregates them. However, current implementations of SA in FL frameworks have limitations, including vulnerability to client dropouts or configuration difficulties. In this paper, we present Salvia, an implementation of SA for Python users in the Flower FL framework. Based on the SecAgg(+) protocols for a semi-honest threat model, Salvia is robust against client dropouts and exposes a flexible and easy-to-use API that is compatible with various machine learning frameworks. We show that Salvia's experimental performance is consistent with SecAgg(+)'s theoretical computation and communication complexities., Comment: Accepted to appear in the 2nd International Workshop on Distributed Machine Learning

Edge computing promises to bring computation close to the end-users to support emergent applications such as virtual reality. However, the computational capacity at the edge of the network is currently limited. To become a pervasive paradigm, edge computing needs highly dispersed decentralized deployments, that, contrary to cloud, cannot benefit from economies of scale. In this situation, crowdsourcing appears attractive - there are plenty of computing devices at the disposal of the general public, and these devices are located exactly where computing power is needed the most - at the edge of the network. Crowdsourcing has been a success maker for scientific computing projects, e.g., SETI@home, or distributed ledger systems empowering decentralized finance. However, as of now, there is no crowdsourced system that addresses the needs of edge computing. In this position paper, we aim to identify the causes of this shortcoming, analyze the potential ways to overcome it, and outline future directions.

This study focuses on sensor classification using machine learning algorithms, to improve data aggregation on the Edge. This aspect is particularly important in large-scale Internet of Things environments, where data aggregation derived from sensors from different vendors often requires human intervention. The proposed research is focused on relying on machine learning to classify sensors based on their semantic descriptions.

Function-as-a-Service or FaaS is a popular delivery model of serverless computing where developers upload code to be executed in the cloud as short running stateless functions. Using smaller functions to decompose processing of larger tasks or workflows introduces the question of how to instrument application control flow to orchestrate an overall task or workflow. In this paper, we examine implications of using different methods to orchestrate the control flow of a serverless data processing pipeline composed as a set of independent FaaS functions. We performed experiments on the AWS Lambda FaaS platform and compared how four different patterns of control flow impact the cost and performance of the pipeline. We investigate control flow using client orchestration, microservice controllers, event-based triggers, and state-machines. Overall, we found that asynchronous methods led to lower orchestration costs, and that event-based orchestration incurred a performance penalty.

In the evasion attacks against deep neural networks (DNN), the attacker generates adversarial instances that are visually indistinguishable from benign samples and sends them to the target DNN to trigger misclassifications. In this paper, we propose a novel multi-view adversarial image detector, namely Argos, based on a novel observation. That is, there exist two "souls" in an adversarial instance, i.e., the visually unchanged content, which corresponds to the true label, and the added invisible perturbation, which corresponds to the misclassified label. Such inconsistencies could be further amplified through an autoregressive generative approach that generates images with seed pixels selected from the original image, a selected label, and pixel distributions learned from the training data. The generated images (i.e., the "views") will deviate significantly from the original one if the label is adversarial, demonstrating inconsistencies that Argos expects to detect. To this end, Argos first amplifies the discrepancies between the visual content of an image and its misclassified label induced by the attack using a set of regeneration mechanisms and then identifies an image as adversarial if the reproduced views deviate to a preset degree. Our experimental results show that Argos significantly outperforms two representative adversarial detectors in both detection accuracy and robustness against six well-known adversarial attacks. Code is available at: https://github.com/sohaib730/Argos-Adversarial_Detection

Recent colorization works implicitly predict the semantic information while learning to colorize black-and-white images. Consequently, the generated color is easier to be overflowed, and the semantic faults are invisible. As a human experience in colorization, our brains first detect and recognize the objects in the photo, then imagine their plausible colors based on many similar objects we have seen in real life, and finally colorize them, as described in the teaser. In this study, we simulate that human-like action to let our network first learn to understand the photo, then colorize it. Thus, our work can provide plausible colors at a semantic level. Plus, the semantic information of the learned model becomes understandable and able to interact. Additionally, we also prove that Instance Normalization is also a missing ingredient for colorization, then re-design the inference flow of U-Net to have two streams of data, providing an appropriate way of normalizing the feature maps from the black-and-white image and its semantic map. As a result, our network can provide plausible colors competitive to the typical colorization works for specific objects., Comment: Code is available at https://minhmanho.github.io/semantic-driven_colorization/

The progressive decentralization of web services brought a need for new instruments to support the development of P2P protocols and applications. The only tools available are network or event simulators, which help the development in artificial or extremely controlled environments. They usually provide low-level support with completely unstructured networks. LibP2P is a library that tries to solve all such problems, letting the users easily deploy nodes on the Internet. It also comes with a set of functionalities, to solve the most important problems of P2P networks, such as NAT traversal, peer and content discovery and routing, and much more. In this paper, we employ LibP2P to implement min-bitcoin, a minimal version of the protocol used by the Bitcoin network to exchange the blocks. We evaluate our implementation to assess the applicability of LibP2P in a real-life blockchain scenario. The evaluation uncovers that setting up communication channels is time-consuming, but data transfers are fast. Additionally, LibP2P efficiently manages cases in which nodes are behind a NAT, under VPN, or in geographically distant places.

International audience; Event streaming and event driven microservices are gaining more attention for architecting scalable cloud software and data systems. Event driven microservices communicate asynchronously over distributed event bus (also called distributed event broker). In event driven microservice architectures a producer microservice creates an event and pushes it into a distributed event broker. On the other side, a consumer microservice pulls the event out of the broker and performs the required business logic on the event. In this work, we propose a methodology and framework for tail latency (high percentile latency) SLA guarantee of event driven consumer microservices. We show a cost-efficient proactive and reactive horizontal autoscale mechanism of consumer microservices to optimize for tail latency of event processing time. In the proposed framework, each latency-sensitive consumer microservice is configured with a maximum latency for event processing, and the framework guarantees autonomous scale in and out of consumer microservice instances to target the required latency SLA under minimal number of replicas. Moreover, we show and discuss the effect of consumer microservices provisioning and synchronization time on the overall latency SLA guarantee.

Path aware networking (PAN) is a promising approach that enables endpoints to participate in end-to-end path selection. PAN unlocks numerous benefits, such as fast failover after link failures, application-based path selection and optimization, and native interdomain multi-path. The utility of PAN hinges on the availability of a large number of high-quality path options. In an inter-domain context, two core questions arise. Can we deploy such an architecture natively in today's Internet infrastructure without creating an overlay relying on BGP? Can we build a scalable multi-path routing system that provides a large number of high-quality paths? We first report on the real-world native deployment of the SCION next-generation architecture, providing a usable PAN infrastructure operating in parallel to today's Internet. We then analyze the scalability of the architecture in an Internet-scale topology. Finally, we introduce a new routing approach to further improve scalability.

Function-as-a-Service (FaaS) has raised a growing interest in how to "tame" serverless to enable domain-specific use cases such as data-intensive applications and machine learning (ML), to name a few. Recently, several systems have been implemented for training ML models. Certainly, these research articles are significant steps in the correct direction. However, they do not completely answer the nagging question of when serverless ML training can be more cost-effective compared to traditional "serverful" computing. To help in this task, we propose MLLess, a FaaS-based ML training prototype built atop IBM Cloud Functions. To boost cost-efficiency, MLLess implements two key optimizations: a significance filter and a scale-in auto-tuner, and leverages them to specialize model training to the FaaS model. Our results certify that MLLess can be 15X faster than serverful ML systems [24] at a lower cost for ML models (such as sparse logistic regression and matrix factorization) that exhibit fast convergence.

Data streaming applications in Cyber-Physical Systems enable high-throughput, low-latency transformations of raw data into value. The performance of such applications, run by Stream Processing Engines (SPEs), can be boosted through custom CPU scheduling. Previous schedulers in the literature require alterations to SPEs to control the scheduling through user-level threads. While such alterations allow for fine-grained control, they hinder the adoption of such schedulers due to the high implementation cost and potential limitations in application semantics (e.g., blocking I/O). Motivated by the above, we explore the feasibility and benefits of custom scheduling without alterations to SPEs but, instead, by orchestrating the OS scheduler (e.g., using nice and cgroup) to enforce the scheduling goals. We propose Lachesis, a standalone scheduling middleware, decoupled from any specific SPE, that can schedule multiple streaming applications, run in one or many nodes, and possibly multiple SPEs. Our evaluation with real-world and synthetic workloads, several SPEs and hardware setups, shows its benefits over default OS scheduling and other state-of-the-art schedulers: up to 75% higher throughput, and 1130x lower average latency once such SPEs reach their peak processing capacity.

In this paper, we present the design and implementation of a privacy preserving event based UTXO (Unspent Transaction Output) transaction system. Unlike the existing approaches that often depend on smart contracts where digital assets are first locked in a vault, and then released according to event triggers, the event based transaction system encodes event outcome as part of the UTXO note and safeguards event privacy by shielding it with zero-knowledge proof based protocols such that associations between UTXO notes and events are hidden from the validators. Without relying on any triggering mechanism, the proposed transaction system separates event processing from the transaction processing where confidential event based UTXO notes (event based UTXOs or conditional UTXOs) can be transferred freely with full privacy in an asynchronous manner, only with their asset values conditional to the linked event outcomes. The main advantage of such design is that it enables free trade of event based digital assets and prevents the assets from being locked. We implemented the proposed transaction system by extending the Zerocoin data model and protocols. The system is implemented and evaluated using xJsnark.

In scientific research, the independent reproduction of experimental results is the source of trust. The release of experimental artifacts enables the reproduction of results; however, additional efforts of researchers are required to prepare and document their experiments accordingly. To honor this increased effort, multiple initiatives were implemented to incentivize the creation and release of experimental artifacts, e.g., awards for papers that provide experimental artifacts. In this work, we want to propose a novel approach toward a reproducible research---a structured experimental workflow that allows the creation of reproducible experiments without requiring additional efforts of the researcher. Moreover, we present our own testbed and toolchain, namely, plain orchestrating service (pos), which enables the creation of such experimental workflows. The experiment is documented by our proposed, fully scripted experiment structure. Further, we consider the entire experimental workflow from experiment orchestration, to data measurement, to result evaluation. In addition, pos provides scripts that enable the automation of the bundling and release of all the created experimental artifacts. In this case study, we release one of our own experiments together with the necessary tools so that others can reproduce our experiment. Additionally, we provide an interactive environment where pos experiments can be executed and reproduced, which is available at https://gallenmu.github.io/pos-artifacts.

This article discusses secure methods to conduct e-voting over a blockchain in three different settings: decentralized voting, centralized remote voting, and centralized polling station voting. These settings cover almost all voting scenarios that occur in practice. A proof-of-concept implementation for decentralized voting over Ethereum’s blockchain is presented. This work demonstrates the suitable use of a blockchain not just as a public bulletin board but, more importantly, as a trustworthy computing platform that enforces the correct execution of the voting protocol in a publicly verifiable manner. We also discuss scaling up a blockchain-based voting application for national elections. We show that for national-scale elections the major verifiability problems can be addressed without having to depend on any blockchain. However, a blockchain remains a viable option to realize a public bulletin board, which has the advantage of being a “preventive” measure to stop retrospective changes on previously published records as opposed to a “detective” measure like the use of mirror websites. CCS Concepts: • Security and privacy

Tactile Internet (TI) refers to the transmission of touch and the real-time control of applications like the remote control for teleoperation of machines, drones or vehicles. Traditional TCP and UDP protocols are not suitable for these applications, because they are mainly designed for content delivery or non real-time interactive applications. Some ideas from transport protocols for real-time multimedia like RTP are closer to Tactile Internet; however, they do not satisfy requirements of low latency and high reliability. In this paper, we present the Multi-connection Tactile Internet Protocol (MTIP), a novel transport protocol on top of the Internet Protocol to support the reliability and latency for Internet applications over multi-homed devices connected to several wireless networks. In order to provide that service, MTIP is based on the combination of sequence numbers and timestamps in packets, enhanced with global clocks and context awareness. The paper focuses on building a state machine based model of MTIP to prove the correct behaviour of the protocol using the SPIN tool to verify a number of relevant correctness properties.

In this paper, we describe the integration of a step-by-step interactive trace table into an existing practice system for introductory Java programming. These autogenerated trace problems provide help and scaffolding for students who have trouble in solving traditional one-step code tracing problems, accommodating a wider variety of learners. Findings from classroom deployments suggest the scaffolding provided by the trace table is a plausible form of help, most notably increases in performance and persistence and lower task difficulty. Based on usage data, we propose future implications for an adaptive version of the interactive trace table based on learner modeling.

The rising demand for data wrangling skills in today’s global market poses new challenges for the programming education community. Non-majors often need to learn it quickly alongside their other subjects. Previous research suggests that subgoal labels offer a powerful scaffolding strategy to help novices decompose problems. Because data wrangling is inherently easy to represent graphically, we wonder whether such labels could be augmented with subgoal graphics. To test this idea, we developed an online tutorial that features subgoal graphics in both programmatic and non-programmatic data wrangling exercises. Following an RCT paradigm, a control group is only given subgoal labels, without any graphics. The platform collects learner activity in order to evaluate the pedagogical benefits. Participants were recruited from multiple institutions (N=197, 134). Our results did not show a significant difference in various learner performance metrics, however subjective feedback from our participants suggest that learners perceive the graphics to be very helpful. We discuss possible reasons for the apparent disparity between objective and subjective data.

While relying on energy harvesting to power Internet of Things (IoT) devices eliminates the maintenance burden of battery replacement, energy generation fluctuation constitutes a major source of uncertainty to design reliable self-powered IoT devices. To characterize spatial-temporal variability of energy harvesting, data acquisition campaigns are needed across the range of potential harvesting sources. In this work we present a dataset to characterize thermal energy sources in residential settings by measuring thermoelectric generator (TEG) operating conditions over 16 deployment locations for periods ranging from 19 to 53 days. We present our easy-to-use thermal energy measurement platform built from off-the-shelf component modules and a custom TEG interface circuit. We demonstrate how the collected measurements can inform the design of energy harvesting IoT devices by deriving the TEG's maximum power output and estimating the available energy at each harvesting location.

Traditional post-hoc high-fidelity scientific visualization (HSV) of numerical simulations requires multiple I/O check-pointing to inspect the simulation progress. The costs of these I/O operations are high and can grow exponentially with increasing problem sizes. In situ HSV dispenses with costly check-pointing I/O operations, but requires additional computing resources to generate the visualization, increasing power and energy consumption. In this paper we present LOOM, a new interweaving approach supported by a task scheduling framework to allow tightly coupled in situ visualization without significantly adding to the overall simulation runtime. The approach exploits the idle times of the numerical simulation threads, due to workload imbalances, to perform the visualization steps. Overall execution time (simulation plus visualization) is minimized. Power requirements are also minimized by sharing the same computational resources among numerical simulation and visualization tasks. We demonstrate that LOOM reduces time to visualization by 3 × compared to a traditional non-interwoven pipeline. Our results here demonstrate good potential for additional gains for large distributed-memory use cases with larger interleaving opportunities.

International audience; We recently introduced the concept of reconfigurable Wireless Networks on Chips (r-WNoCs) for which an on-chip reconfigurable intelligent surface (RIS) endows the wireless on-chip propagation environment with programmability. In this work-in-progress report, we apply this idea to a single-input multiple-output (SIMO) context. Specifically, we demonstrate that using an on-chip RIS we can simultaneously shape multiple channel impulse responses (CIRs) such that they become essentially pulse-like despite rich scattering inside the chip enclosure. Pulse-like CIRs are essential to enable high-speed information exchange between different processors on the same chip with the simple on-off-keying modulation schemes envisaged for WNoCs. © 2021 Owner/Author.

Low-power wireless sensor networks have demonstrated their potential for the detection of rare events such as rockfalls and wildfires, where rapid reporting as well as long-term energy-efficient operation is vital. However, current systems require periodic synchronization to maintain network coordination, heavily rely on node placement or use costly long-range links to infrastructure. We present STeC, a novel wireless communication design that directly exploits the spatial and temporal correlation of signals from the sensed phenomenon to orchestrate event-based communication. We leverage the locality of a co-detection, where a physical event triggers multiple sensors quasi-simultaneously, to efficiently collect, characterize and report sensor data. This eliminates the overhead of periodic network activity and centralized control, resulting in more energy-efficient communication with a lower, more consistent detection latency. In doing so, we propose a fundamentally new approach to avoid the elementary conflict between duty cycle and latency requirements immanent to synchronous protocols by exploiting correlated sensor signals for networking. Experiments using real-world traces of a natural hazard detection application show that STeC reduces the detection latency by up to 87 % compared to standard single-hop communication and outperforms traditional schedule-based methods by up to 58.4 x in energy efficiency., SenSys '21: Proceedings of the 19th ACM Conference on Embedded Networked Sensor Systems, ISBN:978-1-4503-9097-2

Advances in flexible conductive substrates such as conductive wallpaper and paint present new opportunities for optimizing the performance of IoT nodes in smart homes and buildings. In this paper, we explore an unconventional use of such substrates for pulling frequencies of oscillators across IoT devices and wireless front-ends connected to the substrate. We show that by using this technique, we can replace precise crystal oscillators by lower precision and lower cost ceramic oscillators without compromising their ability to be used for tasks that require precise frequencies such as frequency-synchronized multi-static backscatter and synchronized sampling. We present an end-to-end design including a) analysis of conditions under which frequency pulling of oscillators across conductive substrates can work, b) a new technique to detect frequency locking across oscillators without requiring explicit communication, and c) an adaptive method that can be used to synchronize oscillators at minimum power consumption. We then show that these elements can be composed to design a high-performance multi-static backscatter system that performs as well as one that uses a shared high-precision clock but at an order of magnitude less monetary cost. We show that our system can scale and operate at very low power, while having low complexity since it requires no explicit interaction among devices attached to the substrate.

Domain generation algorithms (DGAs) prevent the connection between a botnet and its master from being blocked by generating a large number of domain names. Promising single-data-source approaches have been proposed for separating benign from DGA-generated domains. Collaborative machine learning (ML) can be used in order to enhance a classifier's detection rate, reduce its false positive rate (FPR), and to improve the classifier's generalization capability to different networks. In this paper, we complement the research area of DGA detection by conducting a comprehensive collaborative learning study, including a total of 13,440 evaluation runs. In two real-world scenarios we evaluate a total of eleven different variations of collaborative learning using three different state-of-the-art classifiers. We show that collaborative ML can lead to a reduction in FPR by up to 51.7%. However, while collaborative ML is beneficial for DGA detection, not all approaches and classifier types profit equally. We round up our comprehensive study with a thorough discussion of the privacy threats implicated by the different collaborative ML approaches., Accepted at The 3rd Workshop on Cyber-Security Arms Race (CYSARM '21)

International audience; Many security properties of interest are captured by instrumented semantics that model the functional behavior and the leakage of programs. For several important properties, including cryptographic constant-time (CCT), leakage models are sufficiently abstract that one can define instrumented semantics for high-level and low-level programs. One important goal is then to relate leakage of source programs and leakage of their compilation---this can be used, e.g., to prove preservation of CCT. To simplify this task, we put forward the idea of structured leakage. In contrast to the usual modeling of leakage as a sequence of observations, structured leakage is tightly coupled with the operational semantics of programs. This coupling greatly simplifies the definition of leakage transformers that map the leakage of source programs to leakage of their compilation and yields more precise statements about the preservation of security properties. We illustrate our methods on the Jasmin compiler and prove preservation results for two policies of interest: CCT and cost.

Along with gaining popularity of Real-Time Bidding (RTB) based programmatic advertising, the click farm based invalid traffic, which leverages massive real smartphones to carry out large-scale ad fraud campaigns, is becoming one of the major threats against online advertisement. In this study, we take an initial step towards the detection and large-scale measurement of the click farm based invalid traffic. Our study begins with a measurement on the device's features using a real-world labeled dataset, which reveals a series of features distinguishing the fraudulent devices from the benign ones. Based on these features, we develop EvilHunter, a system for detecting fraudulent devices through ad bid request logs with a focus on clustering fraudulent devices. EvilHunter functions by 1) building a classifier to distinguish fraudulent and benign devices; 2) clustering devices based on app usage patterns; and 3) relabeling devices in clusters through majority voting. EvilHunter demonstrates 97% precision and 95% recall on a real-world labeled dataset. By investigating a super click farm, we reveal several cheating strategies that are commonly adopted by fraudulent clusters. We further reduce the overhead of EvilHunter and discuss how to deploy the optimized EvilHunter in a real-world system. We are in partnership with a leading ad verification company to integrate EvilHunter into their industrial platform.

Zero-knowledge succinct arguments of knowledge (zkSNARKs) enable efficient privacy-preserving proofs of membership for general NP languages. Our focus in this work is on post-quantum zkSNARKs, with a focus on minimizing proof size. Currently, there is a 1000x gap in the proof size between the best pre-quantum constructions and the best post-quantum ones. Here, we develop and implement new lattice-based zkSNARKs in the designated-verifier preprocessing model. With our construction, after an initial preprocessing step, a proof for an NP relation of size 2^20 is just over 16 KB. Our proofs are 10.3x shorter than previous post-quantum zkSNARKs for general NP languages. Compared to previous lattice-based zkSNARKs (also in the designated-verifier preprocessing model), we obtain a 42x reduction in proof size and a 60x reduction in the prover's running time, all while achieving a much higher level of soundness. Compared to the shortest pre-quantum zkSNARKs by Groth (Eurocrypt 2016), the proof size in our lattice-based construction is 131x longer, but both the prover and the verifier are faster (by 1.2x and 2.8x, respectively). Our construction follows the general blueprint of Bitansky et al. (TCC 2013) and Boneh et al. (Eurocrypt 2017) of combining a linear probabilistically checkable proof (linear PCP) together with a linear-only vector encryption scheme. We develop a concretely-efficient lattice-based instantiation of this compiler by considering quadratic extension fields of moderate characteristic and using linear-only vector encryption over rank-2 module lattices.

Scheduler side-channels can leak critical information in real-time systems, thus posing serious threats to many safety-critical applications. The main culprit is the inherent determinism in the runtime timing behavior of such systems, e.g., the (expected) periodic behavior of critical tasks. In this paper, we introduce the notion of "schedule indistinguishability/", inspired by work in differential privacy, that introduces diversity into the schedules of such systems while offering analyzable security guarantees. We achieve this by adding a sufficiently large (controlled) noise to the task schedules in order to break their deterministic execution patterns. An "epsilon-Scheduler" then implements schedule indistinguishability in real-time Linux. We evaluate our system using two real applications: (a) an autonomous rover running on a real hardware platform (Raspberry Pi) and (b) a video streaming application that sends data across large geographic distances. Our results show that the epsilon-Scheduler offers better protection against scheduler side-channel attacks in real-time systems while still maintaining good performance and quality-of-service(QoS) requirements.

Security misconfigurations and neglected updates commonly lead to systems being vulnerable. Especially in the context of websites, we often find pages that were forgotten, that is, they were left online after they served their purpose and never updated thereafter. In this paper, we introduce new methodology to detect such forgotten or orphaned web pages. We combine historic data from the Internet Archive with active measurements to identify pages no longer reachable via a path from the index page, yet stay accessible through their specific URL. We show the efficacy of our approach and the real-world relevance of orphaned web-pages by applying it to a sample of 100,000 domains from the Tranco Top 1M. Leveraging our methodology, we find 1,953 pages on 907 unique domains that are orphaned, some of which are 20 years old. Analyzing their security posture, we find that these pages are significantly ((p < 0.01) using (χ2)) more likely to be vulnerable to cross-site scripting (XSS) and SQL injection (SQLi) vulnerabilities than maintained pages. In fact, orphaned pages are almost ten times as likely to suffer from XSS (19.3%) than maintained pages from a random Internet crawl (2.0%), and maintained pages of websites with some orphans are almost three times as vulnerable (5.9%). Concerning SQLi, maintained pages on websites with some orphans are almost as vulnerable (9.5%) as orphans (10.8%), and both are significantly more likely to be vulnerable than other maintained pages (2.7%). Overall, we see a clear hierarchy: Orphaned pages are the most vulnerable, followed by maintained pages on websites with orphans, with fully maintained sites being least vulnerable. We share an open source implementation of our methodology to enable the reproduction and application of our results in practice.

Amplification Distributed Denial of Service (DDoS) attacks' traffic and harm are at an all-time high. To defend against such attacks, distributed attack mitigation platforms, such as traffic scrubbing centers that operate in peering locations, e.g., Internet Exchange Points (IXP), have been deployed in the Internet over the years. These attack mitigation platforms apply sophisticated techniques to detect attacks and drop attack traffic locally, thus, act as sensors of attacks. However, it has not yet been systematically evaluated and reported to what extent coordination of these views by different platforms can lead to more effective mitigation of amplification DDoS attacks. In this paper, we ask the question: "Is it possible to mitigate more amplification attacks and drop more attack traffic when distributed attack mitigation platforms collaborate?"To answer this question, we collaborate with eleven IXPs that operate in three different regions. These IXPs have more than 2,120 network members that exchange traffic at the rate of more than 11 Terabits per second. We collect network data over six months and analyze more than 120k amplification DDoS attacks. To our surprise, more than 80% of the amplification DDoS are not detected locally, although the majority of the attacks are visible by at least three IXPs. A closer investigation points to the shortcomings, such as the multi-protocol profile of modern amplification attacks, the duration of the attacks, and the difficulty of setting appropriate local attack traffic thresholds that will trigger mitigation. To overcome these limitations, we design and evaluate a collaborative architecture that allows participant mitigation platforms to exchange information about ongoing amplification attacks. Our evaluation shows that it is possible to collaboratively detect and mitigate the majority of attacks with limited exchange of information and drop as much as 90% more attack traffic locally.

International audience; Verifiability is a key requirement for electronic voting. However, the use of cryptographic techniques to achieve it usually requires specialist knowledge to understand; hence voters cannot easily assess the validity of such arguments themselves. To address this, solutions have been proposed using simple tables and checks, which require only simple verification steps with almost no cryptography. This simplicity comes at a cost: numerous verification checks must be made on the tables to ensure their correctness, raising the question whether the success of all the small verification steps entails the overall goal of end-to-end verifiability while preserving vote secrecy. Do the final results reflect the voters' will? Moreover, do the verification steps leak information about the voters' choices? In this paper, we provide mathematical foundations and an associated methodology for defining and proving verifiability and voter privacy for table-based election protocols. We apply them to three case studies: the Eperio protocol, Scantegrity, and Chaum's Random-Sample Election protocol. Our methodology helps us, in all three cases, identify previously unknown problems that allow an election authority to cheat and modify the election outcome. Furthermore, it helps us formulate and verify the corrected versions.

What you change is what you fuzz! In an empirical study of all fuzzer-generated bug reports in OSSFuzz, we found that four in every five bugs have been introduced by recent code changes. That is, 77% of 23k bugs are regressions. For a newly added project, there is usually an initial burst of new reports at 2-3 bugs per day. However, after that initial burst, and after weeding out most of the existing bugs, we still get a constant rate of 3-4 bug reports per week. The constant rate can only be explained by an increasing regression rate. Indeed, the probability that a reported bug is a regression (i.e., we could identify the bug-introducing commit) increases from 20% for the first bug to 92% after a few hundred bug reports. In this paper, we introduce regression greybox fuzzing (RGF) a fuzzing approach that focuses on code that has changed more recently or more often. However, for any active software project, it is impractical to fuzz sufficiently each code commit individually. Instead, we propose to fuzz all commits simultaneously, but code present in more (recent) commits with higher priority. We observe that most code is never changed and relatively old. So, we identify means to strengthen the signal from executed code-of-interest. We also extend the concept of power schedules to the bytes of a seed and introduce Ant Colony Optimization to assign more energy to those bytes which promise to generate more interesting inputs. Our large-scale fuzzing experiment demonstrates the validity of our main hypothesis and the efficiency of regression greybox fuzzing. We conducted our experiments in a reproducible manner within Fuzzbench, an extensible fuzzer evaluation platform. Our experiments involved 3+ CPU-years worth of fuzzing campaigns and 20 bugs in 15 open-source C programs available on OSSFuzz.

In this paper, we investigate the effect of a realism mismatch in the voice and appearance of a photorealistic virtual character in virtual reality. While many studies have investigated voice attributes for robots, not much is known about the effect voice naturalness has on the perception of realistic virtual characters. We conducted an experiment in Virtual Reality (VR) with over two hundred participants investigating the mismatch between realistic appearance and unrealistic voice on the feeling of presence, and the emotional response of the user to the character expressing a strong negative emotion (sadness, guilt). We predicted that the mismatched voice would lower social presence and cause users to have a negative emotional reaction and feelings of discomfort towards the character. We found that the concern for the virtual character was indeed altered by the unnatural voice, though interestingly it did not affect social presence.

We present a method to track a smartphone in VR using a fiducial marker displayed on the screen. Using WebRTC transmission protocol, we capture smartphone touchscreen input and the screen contents, copying them to a virtual representation in VR. We present two Fitts’ law experiments assessing the performance of selecting targets displayed on the virtual smartphone screen using this method. The first compares direct vs. indirect input (i.e., virtual smartphone co-located with the physical smartphone, or not), and reveals there is no difference in performance due to input indirection. The second experiment assesses the influence of input scaling, i.e., decoupling the virtual cursor from the actual finger position on the smartphone screen so as to provide a larger virtual tactile surface. Results indicate a small effect for extreme scale factors. We discuss implications for the use of smartphones as input devices in VR.

Remotely controlled camera drones can support live, dynamic, and interactive virtual tours for travelers to overcome distance, expense, and health barriers. Yet, assigning one drone to one traveler may incur unnecessary waste of resources, and an abundance of concurrent drones raises safety concerns. While sharing the input and output of a single drone among multiple concurrent users can alleviate these limitations, standard control sharing protocols, such as turn-taking, are often inefficient. We present Constellation, a multi-user drone control system that synthesizes diverse user goals and generates efficient flight paths for the group. It supports point-of-interest specification on both static 3D environmental maps and live camera views. The generated paths minimize all users’ total extra waiting time. A web-based study with 16 participants show that Constellation could help groups navigate to their points-of-interest faster in comparison to the turn-taking baseline.

The ability to recognize human partners is an important social skill to build personalized and long-term Human-Robot Interactions (HRI). However, in HRI contexts, unfolding in ever-changing and realistic environments, the identification problem presents still significant challenges. Possible solutions consist of relying on a multimodal approach and making robots learn from their first-hand sensory data. To this aim, we propose a framework to allow robots to autonomously organize their sensory experience into a structured dataset suitable for person recognition during a multiparty interaction. Our results demonstrate the effectiveness of our approach and show that it is a promising solution in the quest of making robots more autonomous in their learning process.

Educational VR may help students by being more engaging or improving retention compared to traditional learning methods. However, a student can get distracted in a VR environment due to stress, mind-wandering, unwanted noise, external alerts, etc. Student eye gaze can be useful for detecting these distraction. We explore deep-learning-based approaches to detect distractions from gaze data. We designed an educational VR environment and trained three deep learning models (CNN, LSTM, and CNN-LSTM) to gauge a student’s distraction level from gaze data, using both supervised and unsupervised learning methods. Our results show that supervised learning provided better test accuracy compared to unsupervised learning methods.

We present Adaptic, a novel ”hybrid” active/passive haptic device that can change shape to act as a proxy for a range of virtual objects in VR. We use Adaptic with haptic retargeting to redirect the user’s hand to provide haptic feedback for several virtual objects in arm’s reach using only a single prop. To evaluate the effectiveness of Adaptic with haptic retargeting, we conducted a within-subjects experiment employing a docking task to compare Adaptic to non-matching proxy objects (i.e., Styrofoam balls) and matching shape props. In our study, Adaptic sat on a desk in front of the user and changed shapes between grasps, to provide matching tactile feedback for various virtual objects placed in different virtual locations. Results indicate that the illusion was convincing: users felt they were manipulating several virtual objects in different virtual locations with a single Adaptic device. Docking performance (completion time and accuracy) with Adaptic was comparable to props without haptic retargeting.

Consumer IoT devices are becoming increasingly popular, with most leveraging TLS to provide connection security. In this work, we study a large number of TLS-enabled consumer IoT devices to shed light on how effectively they use TLS, in terms of establishing secure connections and correctly validating certificates, and how observed behavior changes over time. To this end, we gather more than two years of TLS network traffic from IoT devices, conduct active probing to test for vulnerabilities, and develop a novel black- box technique for exploring the trusted root stores in IoT devices by exploiting a side-channel through TLS Alert Messages. We find a wide range of behaviors across devices, with some adopting best security practices but most being vulnerable in one or more of the following ways: use of old/insecure protocol versions and/or ciphersuites, lack of certificate validation, and poor maintenance of root stores. Specifically, we find that at least 8 IoT devices still include distrusted certificates in their root stores, 11/32 devices are vulnerable to TLS interception attacks, and that many devices fail to adopt modern protocol features over time. Our findings motivate the need for IoT manufacturers to audit, upgrade, and maintain their devices’ TLS implementations in a consistent and uniform way that safeguards all of their network traffic. USA NSF EU H2020 Spanish Ministry of Science Consumer Reports TRUE inpress

Cloud computing has seen continuous growth over the last decade. The recent rise in popularity of next-generation applications brings forth the question: "Can current cloud infrastructure support the low latency requirements of such apps?" Specifically, the interplay of wireless last-mile and investments of cloud operators in setting up direct peering agreements with ISPs globally to current cloud reachability and latency has remained largely unexplored. This paper investigates the state of end-user to cloud connectivity over wireless media through extensive measurements over six months. We leverage 115,000 wireless probes on the Speed-checker platform and 195 cloud regions from 9 well-established cloud providers. We evaluate the suitability of current cloud infrastructure to meet the needs of emerging applications and highlight various hindering pressure points. We also compare our results to a previous study over RIPE Atlas. Our key findings are: (i) the most impact on latency comes from the geographical distance to the datacenter; (ii) the choice of a measurement platform can significantly influence the results; (iii) wireless last-mile access contributes significantly to the overall latency, almost surpassing the impact of the geographical distance in many cases. We also observe that cloud providers with their own private network backbone and direct peering agreements with serving ISPs offer noticeable improvements in latency, especially in its consistency over longer distances.