Input Events Research Articles

Calcium and strontium isotope systematics in the lagoon-estuarine environments of South Australia: Implications for water source mixing, carbonate fluxes and fish migration

This study uses Ca and Sr isotopes (δ44/40Ca and 87Sr/86Sr), coupled with elemental ratios, to better understand the water source apportionment and carbonate output in the Coorong, Lower Lakes and Murray Mouth Estuary, which represents the terminus of Australia’s longest river system. The geochemistry of waters in the Coorong (i.e., North and South Lagoon) can be explained by mixing of three major components, including: (i) the Southern Ocean seawater, (ii) local freshwaters, and (iii) hypersaline lagoon waters, the latter significantly modified by ongoing evaporation and carbonate formation. The Sr and Ca isotope composition of the North Lagoon is indistinguishable from that of the Southern Ocean (i.e., normal salinity of ∼35 PSU), with the exception of transient freshwater input events that can temporarily lower the salinity to brackish levels. Interestingly, our results from the hypersaline South Lagoon (salinity up to ∼120 PSU) confirmed that the latter is highly evaporated brackish water (with ≥40% contribution from continent-derived waters), which has been additionally modified by in-situ carbonate precipitation. Importantly, our Ca isotope and elemental constraints showed that about 15–17% of the dissolved Ca2+ in the South Lagoon has been removed as CaCO3 (primarily as aragonite). This in turn has implications for the local carbonate cycle and blue carbon studies, suggesting that the South Lagoon acts as a net sink for the dissolved inorganic carbon (DIC).Ca isotope data from the otoliths of smallmouth hardyhead fish species (Atherinosoma microstoma) collected in the Coorong indicate that δ44/40Ca is primarily controlled by biological processes (i.e., kinetic isotope fractionation effects related to growth rate), rather than by the Ca isotope composition of local lagoon waters. As to 87Sr/86Sr in otoliths, the latter confirmed the importance of continent-derived water sources in the Coorong, recorded over the life span of the fish.Overall, with suitable fossil carbonate archives (e.g., bivalve shells, foraminifera), our calibration of 87Sr/86Sr and δ44/40Ca in the modern hydrological system, with respect to a large salinity gradient (ranging from fresh to hypersaline, i.e., 0 to ∼120 PSU), implies potential future applications of these isotope tracers in carbonate-producing coastal systems, which include (i) tracing and apportioning different water sources, (ii) quantifying local carbonate outputs, and (iii) reconstructing paleo-salinity changes.

A Parametric EEG Signal Model for BCIs with Rapid-Trial Sequences.

Electroencephalogram (EEG) signals have been shown very effective for inferring user intents in brain-computer interface (BCI) applications. However, existing EEG-based BCIs, in many cases, lack sufficient performance due to utilizing classifiers that operate on EEG signals induced by individual trials. While many factors influence the classification performance, an important aspect that is often ignored is the temporal dependency of these trial-EEG signals, in some cases impacted by interference of brain responses to consecutive target and non-target trials. In this study, the EEG signals are analyzed in a parametric sequence-based fashion, which considers all trials that induce brain responses in a rapid-sequence fashion, including a mixture of consecutive target and non-target trials. EEG signals are described as a linear combination of time-shifted cortical source activities plus measurement noise. Using a superposition of time invariant with an auto-regressive (AR) process, EEG signals are treated as a linear combination of a stationary Gaussian process and time-locked impulse responses to the stimulus (input events) onsets. The model performance is assessed in the framework of a rapid serial visualization presentation (RSVP) based typing task for three healthy subjects across two sessions. Signal modeling in this fashion yields promising performance outcomes considering a single EEG channel to estimate the user intent.

Approaching Retinal Ganglion Cell Modeling and FPGA Implementation for Robotics.

Taking inspiration from biology to solve engineering problems using the organizing principles of biological neural computation is the aim of the field of neuromorphic engineering. This field has demonstrated success in sensor based applications (vision and audition) as well as in cognition and actuators. This paper is focused on mimicking the approaching detection functionality of the retina that is computed by one type of Retinal Ganglion Cell (RGC) and its application to robotics. These RGCs transmit action potentials when an expanding object is detected. In this work we compare the software and hardware logic FPGA implementations of this approaching function and the hardware latency when applied to robots, as an attention/reaction mechanism. The visual input for these cells comes from an asynchronous event-driven Dynamic Vision Sensor, which leads to an end-to-end event based processing system. The software model has been developed in Java, and computed with an average processing time per event of 370 ns on a NUC embedded computer. The output firing rate for an approaching object depends on the cell parameters that represent the needed number of input events to reach the firing threshold. For the hardware implementation, on a Spartan 6 FPGA, the processing time is reduced to 160 ns/event with the clock running at 50 MHz. The entropy has been calculated to demonstrate that the system is not totally deterministic in response to approaching objects because of several bioinspired characteristics. It has been measured that a Summit XL mobile robot can react to an approaching object in 90 ms, which can be used as an attentional mechanism. This is faster than similar event-based approaches in robotics and equivalent to human reaction latencies to visual stimulus.

A memory-bounded, deterministic and terminating semantics for the synchronous programming language Céu

Céu is a synchronous programming language for embedded soft real-time systems. It focuses on control-flow safety features, such as safe shared-memory concurrency and safe abortion of lines of execution, while enforcing memory bounded, deterministic, and terminating reactions to the environment. In this work, we present a small-step structural operational semantics for Céu and a proof that reactions have the properties enumerated above: that for a given arbitrary timeline of input events, multiple executions of the same program always react in bounded time and arrive at the same final finite memory state.

Research of the processing technology for time complex event based on LSTM

With the huge amount of data, it is increasingly meaningful to combine different business system data with potential values. In the traditional event description, the input event flow of the event engine is a single atomic event type. The event predicate constraint contains simple attribute value, comparison operation and simple aggregation operation. The time constraint between events always simply. This makes the traditional detection method cannot meet the requirements such as financial, medical and other relatively accurate time requirements, event predicate constraints require more complex applications. Thus, this paper introduces the long short-term memory network model (LSTM), designs a multivariate event input to process these data based on TCN quantitative timing constraint representation model and predicate constraint representation model. In this paper, an innovative method makes the complex event processing technology more high efficient. By the analysis 200 million records of 2045 stocks, the results show that the processing technology of the complex events is more effective, more efficient.

Developing a dynamic model for risk analysis under uncertainty: Case of third-party damage on subsea pipelines

Third-party damage is an important factor leading to subsea pipelines failure, and risk analysis an efficient approach to mitigate and control such events. However, available crisp probabilities for input events are usually limited, missing or unknown, which introduces data uncertainty. Furthermore, conventional risk analysis methods are known to have a static structure, which introduces model uncertainty. This paper presents a dynamic model for risk analysis under uncertainty and illustrates it by a case of third-party damage on subsea pipelines. Proposed model makes use of fuzzy set theory and evidence theory to handle data uncertainty, and utilizes Bayesian network (BN) to address model uncertainty. Primary accident scenario is developed by the FT-ESD approach, and it is transformed into BN to circumvent model uncertainty by relaxing the limitations of conventional methods. Expert elicitation is integrated into fuzzy set theory and evidence theory to obtain the crisp probabilities of input events in BN. Based on the model, a robust probability reasoning is conducted, through which the most probable factors contributing to the occurrence of unexpected consequence are identified. As new observations become available, potential accident probabilities are updated over time to produce a dynamic risk profile. The case study demonstrates the applicability and effectiveness of the model, which indicates that it is an alternative approach for risk analysis in the process industries under uncertainty.

FEEL: Featured Event Embedding Learning

Statistical script learning is an effective way to acquire world knowledge which can be used for commonsense reasoning. Statistical script learning induces this knowledge by observing event sequences generated from texts. The learned model thus can predict subsequent events, given earlier events. Recent approaches rely on learning event embeddings which capture script knowledge. In this work, we suggest a general learning model–Featured Event Embedding Learning (FEEL)–for injecting event embeddings with fine grained information. In addition to capturing the dependencies between subsequent events, our model can take into account higher level abstractions of the input event which help the model generalize better and account for the global context in which the event appears. We evaluated our model over three narrative cloze tasks, and showed that our model is competitive with the most recent state-of-the-art. We also show that our resulting embedding can be used as a strong representation for advanced semantic tasks such as discourse parsing and sentence semantic relatedness.

AATT+: Effectively manifesting concurrency bugs in Android apps

Smartphones are indispensable in people's daily activities, and smartphone apps tend to be increasingly concurrent due to the wide use of multi-core devices and technologies. Due to this tendency, developers are increasingly unable to tackle the complexity of concurrent apps and to avoid subtle concurrency bugs. To better address this issue, we propose a novel approach to detecting concurrency bugs in Android apps based on the fact that one can generate simultaneous input events and their schedules for an app, which would easily trigger concurrency bugs in an app. We conduct systematic state space exploration to find potentially conflicting resource accesses in an Android app. The app is then automatically pressure-tested by guided event and schedule generation. We implemented our prototype tool named AATT+ and evaluated it with two sets of real-world Android apps. Benchmarking using 15 Android apps with previously known concurrency bugs, AATT+ and existing concurrency-unaware techniques detected 10 and 1 bugs, respectively. Evaluated with another set of 17 popular Android apps, AATT+ detected 11 concurrency bugs and 7 of them were previously unknown, achieving an over 80% higher detection rate than existing concurrency-unaware techniques. • Proposed an effective approach to detecting concurrency bugs in Android apps by combining event and schedule generation. • Implemented a prototype tool AATT+ and evaluated it using real-world Android apps. • Studied concurrency bugs in our experiments and identified common bug patterns.

Prism waves for imaging steep geologic features and sediment terminations against salt flanks: Examples from the Gulf of Mexico

Subsalt imaging in the Gulf of Mexico often suffers from illumination problems where various steep geologic features are not captured with conventional imaging techniques due to the lack of a direct wave path from the source to the steep reflector and back to the receiver. The exact geometry of these invisible salt flanks or salt welds is of great interest to interpreters during salt model building and prospect mapping. This paper discusses a suite of technologies that uses prism waves for imaging geologic interfaces with reverse time migration (RTM). Prism wave RTM imaging represents an example of imaging with multiples, Due to its specific geometric setup, it is often complementary to traditional RTMs. Specifically, the prism wave reflections can illuminate steep features that are invisible otherwise. Furthermore, by utilizing prism wave RTM in “reverse,” one can specify the steep event as the input prism wave generator (often a well-constrained salt boundary) and obtain an alternative (and perhaps cleaner) image of the sediment geometries next to this input event. In the industry, a lot of resources are spent on trying to improve illumination for the geologic features of interest by acquiring additional seismic surveys. With prism wave imaging, one can instead leverage the seismic signal that has already been recorded. We believe that even with moderate prism wave illumination uplift, the interpretive value gained can be quite large. To give the reader a glimpse of what is possible, we present several real data examples highlighting the utility of prism waves in interpretation.

A Configurable Event-Driven Convolutional Node with Rate Saturation Mechanism for Modular ConvNet Systems Implementation.

Convolutional Neural Networks (ConvNets) are a particular type of neural network often used for many applications like image recognition, video analysis or natural language processing. They are inspired by the human brain, following a specific organization of the connectivity pattern between layers of neurons known as receptive field. These networks have been traditionally implemented in software, but they are becoming more computationally expensive as they scale up, having limitations for real-time processing of high-speed stimuli. On the other hand, hardware implementations show difficulties to be used for different applications, due to their reduced flexibility. In this paper, we propose a fully configurable event-driven convolutional node with rate saturation mechanism that can be used to implement arbitrary ConvNets on FPGAs. This node includes a convolutional processing unit and a routing element which allows to build large 2D arrays where any multilayer structure can be implemented. The rate saturation mechanism emulates the refractory behavior in biological neurons, guaranteeing a minimum separation in time between consecutive events. A 4-layer ConvNet with 22 convolutional nodes trained for poker card symbol recognition has been implemented in a Spartan6 FPGA. This network has been tested with a stimulus where 40 poker cards were observed by a Dynamic Vision Sensor (DVS) in 1 s time. Different slow-down factors were applied to characterize the behavior of the system for high speed processing. For slow stimulus play-back, a 96% recognition rate is obtained with a power consumption of 0.85 mW. At maximum play-back speed, a traffic control mechanism downsamples the input stimulus, obtaining a recognition rate above 63% when less than 20% of the input events are processed, demonstrating the robustness of the network.

Multidimensional inverse-scattering series internal multiple prediction in the coupled plane-wave domain

The inverse-scattering series internal multiple prediction and attenuation algorithm predicts multiples using certain combinations of input seismic reflection data events, which are computed in the wavenumber/pseudodepth or plane-wave/vertical traveltime (i.e., [Formula: see text]) domains. Significant differences can arise in the algorithms’ output and computational expense depending on which domain is used. Many of these are traceable to the response of the algorithm to the users’ choice of the search-limiting parameter [Formula: see text]. The question of which domain is optimal can be addressed with benchmark synthetics. The compactness of the input to the plane-wave domain algorithm leads to the expectation that it will have a reduced computational expense. Also, the lack of increase in the dominant period (i.e., the “width”) of input events as the horizontal slowness increases leads to the expectation that it will respond well to a constant [Formula: see text]. Both of these expectations are borne out with a 1.5D benchmark example. A 2D plane-wave prediction requires the data to be transformed to the [Formula: see text], or coupled plane-wave, domain, involving source- and receiver-side horizontal slownesses. An implementation of this transform leads to the first numerical examples of full 2D inverse series [Formula: see text] prediction. The arrival times, relative amplitudes, and moveout patterns of multiples from dipping horizons are seen in a benchmark synthetic example to be faithfully determined in the plane-wave formulation; waveform mismatches are, however, observed, which are traceable to the numerics of the forward and inverse transforms. High-resolution Radon transforms are a good candidate to improve the match.

The tectonic setting and evolution of the 2.7 Ga Kalgoorlie–Kurnalpi Rift, a world-class Archean gold province

The Yilgarn Craton results from three major mantle input events (at ca 3.0–2.9, 2.8 and 2.7 Ga) that have interacted with > 3.0 Ga continental crust. Zircon geochronology and Sm-Nd isotopic data subdivide the craton into an older Yilgarn proto-craton and the younger, more primitive Eastern Goldfields Superterrane (EGST). Formation of the Kalgoorlie-Kurnalpi Rift (KKR) within the EGST was associated with the 2.7 Ga event, which exploited weakened crust at the eastern margin of the Yilgarn proto-craton where thick sequences of komatiite and basalt were erupted between ca 2710 and 2690 Ma in the Kalgoorlie Terrane. Calc-alkaline volcanism in the Kurnalpi Terrane began at ca 2730 Ma and continued to ca 2690 Ma, overlapping rifting and plume-related volcanism in the Kalgoorlie Terrane. Deposition of siliciclastic sedimentary rocks within basins at ca 2660 resulted from an intra-orogenic extensional event and coincided with the transition from High-Ca to Low-Ca granite magmatism and peak emplacement of intrusions with a metasomatised mantle source component. Most aspects of the KKR are satisfied by broadly coincident plume-related magmatism in the Kalgoorlie Terrane and westward subduction to the east of the Burtville Terrane. Geochemical characteristics of 2730–2700 Ma calc-alkaline volcanism and 2685–2630 Ma low-SiO2 and alkali-rich intrusions support models for a continental margin subduction zone setting. World-class gold deposits formed in reactivated margins of the KKR, which became flux zones for mantle-derived magmas, hydrothermal fluids and heat during 2675–2620 Ma orogenesis. The orogenic gold mineralisation can be subdivided into proximal intrusion-related and distal-source deposits.

Time-to-Digital Converter With Sample-and-Hold and Quantization Noise Scrambling Using Harmonics in Ring Oscillators

A high-resolution, high-bandwidth, and noise-scrambling, time-to-digital converter (TDC) is presented. Its architecture, which exploits harmonics in ring oscillators, provides a sample-and-hold mechanism in the form of relative phase. This storage mechanism is highly insensitive to noise and allows for oversampling between input events, therefore, can be designed for very high bandwidth. It can achieve lower quantization noise with fewer measurements than noise-shaping TDCs. This paper presents the architecture in detail, an in-depth analysis of noise sensitivity of the time storage mechanism, and the results from a prototype implemented in a 28-nm CMOS process.

Method for Latency Measurement of Visual Feedback-Based Master–Slave Minimally Invasive Surgical Robot

To measure the latency between human motion stimulation and stereo image display response in a visual feedback-based minimally invasive surgical (MIS) robotic system, a method was proposed by comparing the orientations of input and output events through image-processing technology. This method used a black bar to keep pace with the measured joint rotating at a number of speeds. During tests, an external camera was placed in front of the apparatus with a proper visual field, so that it can simultaneously view orientations of both bars fixed on the corresponding joints. After quantitatively analyzing the accuracy of the proposed measurement method, the method was applied to a visual feedback-based master–slave robotic system with two-degrees-of-freedom. Experimental results show that the latency of the overall system was approximately 250 ms, and the opposite clearance of the measured joint was in the range of 1.7°–1.9°.

Wearable expert system development: definitions, models and challenges for the future

PurposeA wearable expert system (WES) is an expert system designed and implemented to obtain input from and give outputs to wearable devices. Among its distinguishing features are the direct cooperation between domain experts and users, and the interaction with a knowledge maintenance system devoted to dynamically update the knowledge base taking care of the evolving scenario. The paper aims to discuss these issues.Design/methodology/approachThe WES development method is based on the Knowledge Acquisition Framework based on Knowledge Artifact (KAFKA) framework. KAFKA employs multiple knowledge artifacts, each devoted to the acquisition and management of a specific kind of knowledge. The KAFKA framework is introduced from both the conceptual and computational points of view. An example is given which demonstrates the interaction, within this framework, of taxonomies, Bayesian networks and rule-based systems. An experimental assessment of the framework usability is also given.FindingsThe most interesting characteristic of WESs is their capability to evolve over time, due both to the measurement of new values for input variables and to the detection of new input events, that can be used to modify, extend and maintain knowledge bases and to represent domains characterized by variability over time.Originality/valueWES is a new and challenging concept, dealing with the possibility for a user to develop his/her own decision support systems and update them according to new events when they arise from the environment. The system fully supports domain experts and users with no particular skills in knowledge engineering methodologies, to create, maintain and exploit their expert systems, everywhere and when necessary.

Integration of Semantic and Episodic Memories.

This paper describes the integration of semantic and episodic memory (EM) models and the benefits of such integration. Semantic memory (SM) is used as a foundation of knowledge and concept learning, and is needed for the operation of any cognitive system. EM retains personal experiences stored based on their significance-it is supported by the SM, and in return, it supports SM operations. Integrated declarative memories are critical for cognitive system development, yet very little research has been done to develop their computational models. We considered structural self-organization of both semantic and episodic memories with a symbolic representation of input events. Sequences of events are stored in EM and are used to build associations in SM. We demonstrated that integration of semantic and episodic memories improves the native operation of both types of memories. Experimental results are presented to illustrate how the two memories complement each other by improving recognition, prediction, and context-based generalization of individual memories.

New parallel processing strategies in complex event processing systems with data streams

Sensor network–based application has gained increasing attention where data streams gathered from distributed sensors need to be processed and analyzed with timely responses. Distributed complex event processing is an effective technology to handle these data streams by matching of incoming events to persistent pattern queries. Therefore, a well-managed parallel processing scheme is required to improve both system performance and the quality-of-service guarantees of the system. However, the specific properties of pattern operators increase the difficulties of implementing parallel processing. To address this issue, a new parallelization model and three parallel processing strategies are proposed for distributed complex event processing systems. The effects of temporal constraints, for example, sliding windows, are included in the new parallelization model to enable the processing load for the overlap between windows of a batch induced by each input event to be shared by the downstream machines to avoid events that may result in wrong decisions. The proposed parallel strategies can keep the complex event processing system working stably and continuously during the elapsed time. Finally, the application of our work is demonstrated using experiments on the StreamBase system regardless of the increased input rate of the stream or the increased time window size of the operator.

DEMATERIALIZZAZIONE DELLA FISICA

Quantum Mechanics has taught us a lesson that goes well beyond a set of new physical laws. It has provided us with genuine “theorems of epistemology”. We now know that there are situations when the observation of the world cannot be considered as the mere reading of a reality pre-existing the observation, whereas there are circumstances where the knowledge of the whole do not corresponds to the knowledge of the parts (holism). We know that there are incompatible properties that are complementary. We know that there exist properties of the whole that are incompatible with any property of each part. The notion of “object” defined in terms of its “properties ” is in contrast with its mereologic connotation according to which objects can be composed to form new objects. The particle ontology as localizable unity is in contrast with the Malament theorem in quantum field theory. To the above add the fact that the two more general fundamental theories in physics – the quantum theory of fields and the general relativity of Einstein – are logically incompatible. In order to reconcile the logical coherence of the observations with their theoretical explanation we need a radical change of paradigm. The solution here proposed is to abandon the vision of the world as a “mechanism” and to substitute it with that of “algorithm”. This is the paradigm of the “universe as a huge computer” that have been latent within the community since Richard Feynman, and nowadays is resurrecting in physics, showing its full theoretical power. In the new algorithmic vision the “mechanics” becomes an emergent phenomenology. Without the need of physical primitive, the new “informational” program allows us to found physics on solid axiomatic grounds. The quantum theory of abstract systems along with the free quantum field theory are derived from information-theoretic axioms. The axioms of quantum theory all have an epistemological connotation, and pertain the possibility of falsifying the propositions of the theory. They reconnect holism with reductionism, probabilism and falsifiability, substituting the notion of “object” with that of “system” and of “event”, and logically implying the theorem of quantum theory without the need of using the abstract Hilbert-spaces toolkit. The free quantum field theory (Weyl, Dirac e Maxwell) is obtained by adding axioms of minimization of algorithmic complexity. The informational framework well separates the notions of “experiment”and “theory”: the theory connects input with output, the experiment being identified with the collection of input, output, and objective intermediate events. The objective reality (experiment) is made of “icons” with which we interact: the theory is the underlying algorithm. The paradigm describes the tapestry of reality as pure software, “software”. “Software without hardware”: the dematerialization of physics. “Reality without realism” means pure logical coherentism.

Mechanistic modelling of a recombinase‐based two‐input temporal logic gate

Site-specific recombinases (SSRs) mediate efficient manipulation of DNA sequences in vitro and in vivo. In particular, serine integrases have been identified as highly effective tools for facilitating DNA inversion, enabling the design of genetic switches that are capable of turning the expression of a gene of interest on or off in the presence of a SSR protein. The functional scope of such circuitry can be extended to biological Boolean logic operations by incorporating two or more distinct integrase inputs. To date, mathematical modelling investigations have captured the dynamical properties of integrase logic gate systems in a purely qualitative manner, and thus such models are of limited utility as tools in the design of novel circuitry. Here, the authors develop a detailed mechanistic model of a two-input temporal logic gate circuit that can detect and encode sequences of input events. Their model demonstrates quantitative agreement with time-course data on the dynamics of the temporal logic gate, and is shown to subsequently predict dynamical responses relating to a series of induction separation intervals. The model can also be used to infer functional variations between distinct integrase inputs, and to examine the effect of reversing the roles of each integrase on logic gate output.

Joint inversion for effective anisotropic velocity model and event locations using S-wave splitting measurements from downhole microseismic data

An accurate velocity model is essential in microseismic imaging. Layered sediments and subvertical cracks from tectonic stress and hydraulic fracturing make velocity models more complicated than isotropy or simple anisotropy, such as vertical transverse isotropy. Downhole microseismic acquisition usually cannot achieve sufficient ray coverage that is required to develop a low-symmetry anisotropic model from traveltimes alone. To solve this problem, we have developed a new type of data, S-wave splitting parameters (the delay time of the slow S-wave and fast S-wave polarization direction), to determine anisotropic models. A genetic algorithm inversion is adopted to solve for the locations of events and the stiffness tensor of an anisotropic medium simultaneously. We applied this method to synthetic waveforms from numerical modeling and successfully recovered the input event locations and the velocity model. The effectiveness of this method is further demonstrated by using real microseismic data acquired in the Bakken shale reservoir. Compared with the inversion with an isotropic velocity model, event locations from an anisotropic model become well-aligned with natural fractures in the Bakken Formation. Our experiments have evidenced that adding full S-wave splitting parameters makes a significant improvement in constraining a low-symmetry anisotropic model from downhole microseismic data.