Abstract: Recently there has been a significant amount of research on developing consensus based algorithms for distributed optimization motivated by applications that vary from large scale machine learning to wireless sensor networks. This work describes and proves convergence of a new algorithm called Push-Sum Distributed Dual Averaging which combines a recent optimization algorithm [1] with a push-sum consensus protocol [2]. As we discuss, the use of push-sum has significant advantages. Restricting to doubly stochastic consensus protocols is not required and convergence to the true average consensus is guaranteed without knowing the stationary distribution of the update matrix in advance. Furthermore, the communication semantics of just summing the incoming information make this algorithm truly asynchronous and allow a clean analysis when varying intercommunication intervals and communication delays are modelled. We include experiments in simulation and on a small cluster to complement the theoretical analysis.

Abstract: This paper discusses practical consensus-based distributed optimization algorithms. In consensus-based optimization algorithms, nodes interleave local gradient descent steps with consensus iterations. Gradient steps drive the solution to a minimizer, while the consensus iterations synchronize the values so that all nodes converge to a network-wide optimum when the objective is convex and separable. The consensus update requires communication. If communication is synchronous and nodes wait to receive one message from each of their neighbors before updating then progress is limited by the slowest node. To be robust to failing or stalling nodes, asynchronous communications should be used. Asynchronous protocols using bi-directional communications cause deadlock, and so one-directional protocols are necessary. However, with one-directional asynchronous protocols it is no longer possible to guarantee the consensus matrix is doubly stochastic. At the same time it is essential that the coordination protocol achieve consensus on the average to avoid biasing the optimization objective. We report on experiments running Push-Sum Distributed Dual Averaging for convex optimization in a MPI cluster. The experiments illustrate the benefits of using asynchronous consensus-based distributed optimization when some nodes are unreliable and may fail or when messages experience time-varying delays.

Abstract: Consider a set of $n>2$ identical mobile computational entities in the plane, called robots, operating in Look-Compute-Move cycles, without any means of direct communication. The Gathering Problem is the primitive task of all entities gathering in finite time at a point not fixed in advance, without any external control. The problem has been extensively studied in the literature under a variety of strong assumptions (e.g., synchronicity of the cycles, instantaneous movements, complete memory of the past, common coordinate system, etc.). In this paper we consider the setting without those assumptions, that is, when the entities are oblivious (i.e., they do not remember results and observations from previous cycles), disoriented (i.e., have no common coordinate system), and fully asynchronous (i.e., no assumptions exist on timing of cycles and activities within a cycle). The existing algorithmic contributions for such robots are limited to solutions for $n \leq 4$ or for restricted sets of initial configura...

Abstract: A system and method for providing an augmented reality environment in which the environmental mapping process is decoupled from the localization processes performed by one or more mobile devices is described. In some embodiments, an augmented reality system includes a mapping system with independent sensing devices for mapping a particular real-world environment and one or more mobile devices. Each of the one or more mobile devices utilizes a separate asynchronous computing pipeline for localizing the mobile device and rendering virtual objects from a point of view of the mobile device. This distributed approach provides an efficient way for supporting mapping and localization processes for a large number of mobile devices, which are typically constrained by form factor and battery life limitations.

Abstract: P2012 is an area- and power-efficient many-core computing accelerator based on multiple globally asynchronous, locally synchronous processor clusters. Each cluster features up to 16 processors with independent instruction streams sharing a multi-banked one-cycle access L1 data memory, a multi-channel DMA engine and specialized hardware for synchronization and aggressive power management. P2012 is 3D stacking ready and can be customized to achieve extreme area and energy efficiency by adding domain-specific HW IPs to the cluster. The first P2012 SoC prototype in 28nm CMOS will sample in Q3, featuring four 16-processor clusters, a 1MB L2 memory and delivering 80GOPS (with 32 bit single precision floating point support) in 18mm2 with 2W power consumption (worst-case). P2012 can run standard OpenCL™ and proprietary Native Programming Model SW components to achieve the highest level of control on application-to-resource mapping. A dedicated version of the OpenCV vision library is provided in the P2012 SW Development Kit to enable visual analytics acceleration. This paper will discuss preliminary performance measurements of common feature extraction and tracking algorithms, parallelized on P2012, versus sequential execution on ARM CPUs.

Abstract: We present a novel event-based stereo matching algorithm that exploits the asynchronous visual events from a pair of silicon retinas. Unlike conventional frame-based cameras, recent artificial retinas transmit their outputs as a continuous stream of asynchronous temporal events, in a manner similar to the output cells of the biological retina. Our algorithm uses the timing information carried by this representation in addressing the stereo-matching problem on moving objects. Using the high temporal resolution of the acquired data stream for the dynamic vision sensor, we show that matching on the timing of the visual events provides a new solution to the real-time computation of 3-D objects when combined with geometric constraints using the distance to the epipolar lines. The proposed algorithm is able to filter out incorrect matches and to accurately reconstruct the depth of moving objects despite the low spatial resolution of the sensor. This brief sets up the principles for further event-based vision processing and demonstrates the importance of dynamic information and spike timing in processing asynchronous streams of visual events.

Abstract: Inference for probabilistic graphical models is still very much a practical challenge in large domains. The commonly used and effective belief propagation (BP) algorithm and its generalizations often do not converge when applied to hard, real-life inference tasks. While it is widely recognized that the scheduling of messages in these algorithms may have significant consequences, this issue remains largely unexplored. In this work, we address the question of how to schedule messages for asynchronous propagation so that a fixed point is reached faster and more often. We first show that any reasonable asynchronous BP converges to a unique fixed point under conditions similar to those that guarantee convergence of synchronous BP. In addition, we show that the convergence rate of a simple round-robin schedule is at least as good as that of synchronous propagation. We then propose residual belief propagation (RBP), a novel, easy-to-implement, asynchronous propagation algorithm that schedules messages in an informed way, that pushes down a bound on the distance from the fixed point. Finally, we demonstrate the superiority of RBP over state-of-the-art methods for a variety of challenging synthetic and real-life problems: RBP converges significantly more often than other methods; and it significantly reduces running time until convergence, even when other methods converge.

Abstract: Micromanipulation systems have recently been receiving increased attention. Teleoperated or automated micromanipulation is a challenging task due to the need for high-frequency position or force feedback to guarantee stability. In addition, the integration of sensors within micromanipulation platforms is complex. Vision is a commonly used solution for sensing; unfortunately, the update rate of the frame-based acquisition process of current available cameras cannot ensure-at reasonable costs-stable automated or teleoperated control at the microscale level, where low inertia produces highly unreachable dynamic phenomena. This paper presents a novel vision-based microrobotic system combining both an asynchronous address event representation silicon retina and a conventional frame-based camera. Unlike frame-based cameras, recent artificial retinas transmit their outputs as a continuous stream of asynchronous temporal events in a manner similar to the output cells of a biological retina, enabling high update rates. This paper introduces an event-based iterative closest point algorithm to track a microgripper's position at a frequency of 4 kHz. The temporal precision of the asynchronous silicon retina is used to provide a haptic feedback to assist users during manipulation tasks, whereas the frame-based camera is used to retrieve the position of the object that must be manipulated. This paper presents the results of an experiment on teleoperating a sphere of diameter around 50 μm using a piezoelectric gripper in a pick-and-place task.

Abstract: The purpose of our research was to examine the influence of an online protocol on asynchronous discussions. A mixed-methods study compared two online graduate classes: one that used a protocol and one that did not use a protocol for the same discussion about a complex reading. Analysis of the data revealed that the online protocol more evenly distributed the presence of cognitive, social, and teaching elements necessary to create and sustain an online community of inquiry. Use of the protocol also promoted more shared group cognition and more student ownership of the discussion and empowered students to facilitate themselves, helping to reduce the instructor workload. These findings may enable educators to provide more dynamic interaction and richer learning experiences in asynchronous online environments.

Abstract: A computing device may receive an incoming communication and, in response, generate a notification that indicates that the incoming communication can be accessed using a particular application on the communication device. The computing device may further provide an audio signal indicative of the notification and automatically activate a listening mode. The computing device may receive a voice input during the listening mode, and an input text may be obtained based on speech recognition performed upon the voice input. A command may be detected in the input text. In response to the command, the computing device may generate an output text that is based on at least the notification and provide a voice output that is generated from the output text via speech synthesis. The voice output identifies at least the particular application.

Abstract: Spiking neural P systems (SN P systems, for short) are a class of distributed parallel computing devices inspired from the way neurons communicate by means of spikes. Asynchronous SN P systems are non-synchronized systems, where the use of spiking rules (even if they are enabled by the contents of neurons) is not obligatory. It remains open whether asynchronous SN P systems with standard spiking rules are equivalent with Turing machines. In this paper, with a biological inspiration (in order to achieve some specific biological functioning, neurons from the same functioning motif or community work synchronously to cooperate with each other), we introduce the notion of local synchronization into asynchronous SN P systems. The computation power of asynchronous SN P systems with local synchronization is investigated. Such systems consisting of general neurons (respectively, unbounded neurons) and using standard spiking rules are proved to be universal. Asynchronous SN P systems with local synchronization consisting of bounded neurons and using standard spiking rules characterize the semilinear sets of natural numbers. These results show that the local synchronization is useful, it provides some ''programming capacity'' useful for achieving a desired computation power.

Abstract: The Internet is increasingly used as a tool in qualitative research. In particular, asynchronous online focus groups are used when factors such as cost, time, or access to participants can make conducting face-to-face research difficult. In this article we consider key methodological issues involved in using asynchronous online focus groups to explore experiences of health and illness. The written nature of Internet communication, the lack of physical presence, and the asynchronous, longitudinal aspects enable participants who might not normally contribute to research studies to reflect on their personal stories before disclosing them to the researcher. Implications for study design, recruitment strategies, and ethics should be considered when deciding whether to use this method.

Abstract: Summary This paper presents a new high speed vision system using an asynchronous address-event representation camera. Within this framework, an asynchronous event-based real-time Hough circle transform is developed to track microspheres. Thetechnologypresentedinthispaperallowsforarobustrealtimeevent-basedmultiobjectpositiondetectionatafrequency of several kHz with a low computational cost. Brownian motion is also detected within this context with both high speed and precision. The carried-out work is adapted to the automated or remote-operated microrobotic systems fulfilling theirneedofanextremelyfastvisionfeedback.Itisalsoavery promising solution tothemicrophysical phenomena analysis and particularly for the micro/nanoscale force measurement.

Abstract: Methods and apparatus, including computer program products, implementing and using techniques for selectively accessing one or more web services from a client machine, the one or more web services and the client machine being accessible over a network. A request for information is received from a client machine with a conversion engine. The request is received over a synchronous interface. The request is processed in the conversion engine. The processed request is transmitted over an asynchronous interface from the conversion engine to at least one web service. Methods and apparatus, including computer program products, for converting a web service description language file for a synchronous web service into a web service description language file for an asynchronous client machine are also described.

Abstract: In this paper, we propose a distributed asynchronous clock synchronization (DCS) protocol for Delay Tolerant Networks (DTNs). Different from existing clock synchronization protocols, the proposed DCS protocol can achieve global clock synchronization among mobile nodes within the network over asynchronous and intermittent connections with long delays. Convergence of the clock values can be reached by compensating for clock errors using mutual relative clock information that is propagated in the network by contacted nodes. The level of clock accuracy is depreciated with respect to time in order to account for long delays between contact opportunities. Mathematical analysis and simulation results for various network scenarios are presented to demonstrate the convergence and performance of the DCS protocol. It is shown that the DCS protocol can achieve faster clock convergence speed and, as a result, reduces energy cost by half for neighbor discovery.

Abstract: E-learning is among the most important explosion propelled by the internet transformation. This allows users to fruitfully gather knowledge and education both by synchronous and asynchronous methodologies to effectively face the need to rapidly acquire up to date know-how within productive environments. This review paper discusses on e-learning methodologies and tools. The different categories of e-learning that includes informal and blending learning, network and work-based learning. The main focus of e-learning methodologies is on both asynchronous and synchronous methodology. The paper also looked into the three major e-learning tools which are (i) curriculum tools (ii) digital library tools and (iii) knowledge representation tools. The paper resolves that e-learning is a revolutionary way to empower workforce with the skill and knowledge it needs to turn change to an advantage. Consequently, many corporations are discovering that e-learning can be used as a tool for knowledge management. Finally the paper suggests that synchronous tools should be integrated into asynchronous environments to allow for “any-time” learning model. This environment would be primarily asynchronous with background discussion, assignments and assessment taking place and managed through synchronous tools.

Abstract: In this paper we propose an approach to the implementation of controllers with decentralized strategies triggering controller updates. We consider set-ups with a central node in charge of the computation of the control commands, and a set of not co-located sensors providing measurements to the controller node. The solution we propose does not require measurements from the sensors to be synchronized in time. The sensors in our proposal provide measurements in an aperiodic way triggered by local conditions. Furthermore, in the proposed implementation (most of) the communication between nodes requires only the exchange of one bit of information (per controller update), which could aid in reducing transmission delays and as a secondary effect result in fewer transmissions being triggered.

Abstract: In this article, we investigate the problem of controlling node sleep intervals so as to achieve the min-max energy fairness in asynchronous duty-cycling sensor networks. We propose a mathematical model to describe the energy efficiency of such networks and observe that traditional sleep interval setting strategies, for example, operating sensor nodes with an identical sleep interval, or intuitive control heuristics, for example, greedily increasing sleep intervals of sensor nodes with high energy consumption rates, hardly perform well in practice. There is an urgent need to develop an efficient sleep interval control strategy for achieving fair and high energy efficiency. To this end, we theoretically formulate the Sleep Interval Control (SIC) problem and find out that it is a convex optimization problem. By utilizing the convex property, we decompose the original problem and propose a distributed algorithm, called GDSIC. In GDSIC, sensor nodes can tune sleep intervals through a local information exchange such that the maximum energy consumption rate of the network approaches to be minimized. The algorithm is self-adjustable to the traffic load variance and is able to serve as a unified framework for a variety of asynchronous duty-cycling MAC protocols. We implement our approach in a prototype system and test its feasibility and applicability on a 50-node testbed. We further conduct extensive trace-driven simulations to examine the efficiency and scalability of our algorithm with various settings.

Abstract: A system includes an infrastructure as a service (IaaS) layer to provide a set of infrastructure services to manage computing resources in a cloud computing environment. The system includes a service framework layer, separate from the IaaS layer, the service framework layer comprising an asynchronous interface to communicate with the IaaS layer and to create a framework service to extend the IaaS layer in response to a command from an upper tier service, the framework service to employ the asynchronous interface to utilize an infrastructure service from the set of infrastructure services.

Abstract: In this paper, we propose two distributed linear convolutional space-time coding (DLC-STC) schemes for full-duplex (FD) asynchronous cooperative communications. The DLC-STC Scheme 1 is for the case of the complete loop channel cancellation, which achieves the full asynchronous cooperative diversity. The DLC-STC Scheme 2 is for the case of the partial loop channel cancellation and amplifying, where some loop signals are used as the self-coding instead of treated as interference to be directly cancelled. We show this scheme can achieve full asynchronous cooperative diversity. We then evaluate the performance of the two schemes when loop channel information is not accurate and present an amplifying factor control method for the DLC-STC Scheme 2 to improve its performance with inaccurate loop channel information. Simulation results show that the DLC-STC Scheme 1 outperforms the DLC-STC Scheme 2 and the delay diversity scheme if perfect or high quality loop channel information is available at the relay, while the DLC-STC Scheme 2 achieves better performance if the loop channel information is imperfect.

Abstract: The feasibility of a full-duplex physical layer, where a node can transmit and receive at the same time in the same frequency, has been established in recent work. Implicitly, most schemes to-date have assumed synchronous operation which allows clean training of all relevant channels. In this paper, we establish the feasibility of asynchronous full-duplex communication. We show that both modes of asynchronous full-duplex can be enabled: (i) where the start of transmission of a packet at a node precedes start of the reception from the same node, and (ii) where the start of reception precedes the start of transmission. We show that the former mode has a better performance, which is in fact comparable to the performance of synchronous full-duplex. Finally, we also show that the active suppression can be complemented by passive suppression with the use of optimal antenna placment on actual devices. An overall suppression of 80dB is shown possible in actual experiments making full-duplex feasible for realistic deployments.

Abstract: Sequential sampling strategies based on Gaussian processes are now widely used for the optimization of problems involving costly simulations. But Gaussian processes can also generate parallel optimization strategies. We focus here on a new, parameter free, parallel expected improvement criterion for asynchronous optimization. An estimation of the criterion, which mixes Monte Carlo sampling and analytical bounds, is proposed. Logarithmic speed-ups are measured on 1 and 9 dimensional functions.