Search

Declarative process specifications define the behavior of processes by means of rules based on Linear Temporal Logic on Finite Traces (LTLf). In a mining context, these specifications are inferred from, and checked on, multi-sets of runs recorded by information systems (namely, event logs). To this end, being able to gauge the degree to which process data comply with a specification is key. However, existing mining and verification techniques analyze the rules in isolation, thereby disregarding their interplay. In this paper, we introduce a framework to devise probabilistic measures for declarative process specifications. Thereupon, we propose a technique that measures the degree of satisfaction of specifications over event logs. To assess our approach, we conduct an evaluation with real-world data, evidencing its applicability in discovery, checking, and drift detection contexts.

Business Process Simulation (BPS) is an approach to analyze the performance of business processes under different scenarios. For example, BPS allows us to estimate what would be the cycle time of a process if one or more resources became unavailable. The starting point of BPS is a process model annotated with simulation parameters (a BPS model). BPS models may be manually designed, based on information collected from stakeholders and empirical observations, or automatically discovered from execution data. Regardless of its origin, a key question when using a BPS model is how to assess its quality. In this paper, we propose a collection of measures to evaluate the quality of a BPS model w.r.t. its ability to replicate the observed behavior of the process. We advocate an approach whereby different measures tackle different process perspectives. We evaluate the ability of the proposed measures to discern the impact of modifications to a BPS model, and their ability to uncover the relative strengths and weaknesses of two approaches for automated discovery of BPS models. The evaluation shows that the measures not only capture how close a BPS model is to the observed behavior, but they also help us to identify sources of discrepancies., Comment: Preprint submitted to the 21st International Conference on Business Process Management (BPM 2023)

By connecting the LU factorization and the Gram-Schmidt orthogonalization without any normalization, closed-forms for the coefficients of the ordinary least squares estimates are presented. Instead of using matrix inversion explicitly, each of the coefficients is expressed and computed directly as a linear combination of non-normalized Gram-Schmidt vectors and the original data matrix and also in terms of the upper triangular factor from LU factorization. The coefficients may computed iteratively using backward or forward algorithms given.

Process discovery studies ways to use event data generated by business processes and recorded by IT systems to construct models that describe the processes. Existing discovery algorithms are predominantly concerned with constructing process models that represent the control flow of the processes. Agent system mining argues that business processes often emerge from interactions of autonomous agents and uses event data to construct models of the agents and their interactions. This paper presents and evaluates Agent Miner, an algorithm for discovering models of agents and their interactions from event data composing the system that has executed the processes which generated the input data. The conducted evaluation using our open-source implementation of Agent Miner and publicly available industrial datasets confirms that our algorithm can provide insights into the process participants and their interaction patterns and often discovers models that describe the business processes more faithfully than process models discovered using conventional process discovery algorithms.

In general, convolutional neural networks (CNNs) are easy to train, but their essential properties, such as generalization error and adversarial robustness, are hard to control. Recent research demonstrated that singular values of convolutional layers significantly affect such elusive properties and offered several methods for controlling them. Nevertheless, these methods present an intractable computational challenge or resort to coarse approximations. In this paper, we offer a principled approach to alleviating constraints of the prior art at the expense of an insignificant reduction in layer expressivity. Our method is based on the tensor-train decomposition; it retains control over the actual singular values of convolutional mappings while providing structurally sparse and hardware-friendly representation. We demonstrate the improved properties of modern CNNs with our method and analyze its impact on the model performance, calibration, and adversarial robustness. The source code is available at: https://github.com/WhiteTeaDragon/practical_svd_conv, Comment: Published as a conference paper at NeurIPS 2022

Problem Definition: Allocating sufficient capacity to cloud services is a challenging task, especially when demand is time-varying, heterogeneous, contains batches, and requires multiple types of resources for processing. In this setting, providers decide whether to reserve portions of their capacity to individual job classes or to offer it in a flexible manner. Methodology/results: In collaboration with Huawei Cloud, a worldwide provider of cloud services, we propose a heuristic policy that allocates multiple types of resources to jobs and also satisfies their pre-specified service level agreements (SLAs). We model the system as a multi-class queueing network with parallel processing and multiple types of resources, where arrivals (i.e., virtual machines and containers) follow time-varying patterns and require at least one unit of each resource for processing. While virtual machines leave if they are not served immediately, containers can join a queue. We introduce a diffusion approximation of the offered load of such system and investigate its fidelity as compared to the observed data. Then, we develop a heuristic approach that leverages this approximation to determine capacity levels that satisfy probabilistic SLAs in the system with fully flexible servers. Managerial Implications: Using a data set of cloud computing requests over a representative 8-day period from Huawei Cloud, we show that our heuristic policy results in a 20% capacity reduction and better service quality as compared to a benchmark that reserves resources. In addition, we show that the system utilization induced by our policy is superior to the benchmark, i.e., it implies less idling of resources in most instances. Thus, our approach enables cloud operators to both reduce costs and achieve better performance.

In this paper, we analyze how well a machine can solve a general problem in queueing theory. To answer this question, we use a deep learning model to predict the stationary queue-length distribution of an $M/G/1$ queue (Poisson arrivals, general service times, one server). To the best of our knowledge, this is the first time a machine learning model is applied to a general queueing theory problem. We chose $M/G/1$ queue for this paper because it lies "on the cusp" of the analytical frontier: on the one hand exact solution for this model is available, which is both computationally and mathematically complex. On the other hand, the problem (specifically the service time distribution) is general. This allows us to compare the accuracy and efficiency of the deep learning approach to the analytical solutions. The two key challenges in applying machine learning to this problem are (1) generating a diverse set of training examples that provide a good representation of a "generic" positive-valued distribution, and (2) representations of the continuous distribution of service times as an input. We show how we overcome these challenges. Our results show that our model is indeed able to predict the stationary behavior of the $M/G/1$ queue extremely accurately: the average value of our metric over the entire test set is $0.0009$. Moreover, our machine learning model is very efficient, computing very accurate stationary distributions in a fraction of a second (an approach based on simulation modeling would take much longer to converge). We also present a case-study that mimics a real-life setting and shows that our approach is more robust and provides more accurate solutions compared to the existing methods. This shows the promise of extending our approach beyond the analytically solvable systems (e.g., $G/G/1$ or $G/G/c$).

The GlueX experiment at Jefferson Lab ran with its first commissioning beam in late 2014 and the spring of 2015. Data were collected on both plastic and liquid hydrogen targets, and much of the detector has been commissioned. All of the detector systems are now performing at or near design specifications and events are being fully reconstructed, including exclusive production of $\pi^{0}$, $\eta$ and $\omega$ mesons. Linearly-polarized photons were successfully produced through coherent bremsstrahlung and polarization transfer to the $\rho$ has been observed., Comment: 8 pages, 6 figures, Invited contribution to the Hadron 2015 Conference, Newport News VA, September 2015

Results are presented for the first measurement of the double-polarization helicity asymmetry E for the $\eta$ photoproduction reaction $\gamma p \rightarrow \eta p$. Data were obtained using the FROzen Spin Target (FROST) with the CLAS spectrometer in Hall B at Jefferson Lab, covering a range of center-of-mass energy W from threshold to 2.15 GeV and a large range in center-of-mass polar angle. As an initial application of these data, the results have been incorporated into the J\"ulich model to examine the case for the existence of a narrow $N^*$ resonance between 1.66 and 1.70 GeV. The addition of these data to the world database results in marked changes in the predictions for the E observable using that model. Further comparison with several theoretical approaches indicates these data will significantly enhance our understanding of nucleon resonances.

The beam-spin asymmetry, $\Sigma$, for the reaction $\gamma d\rightarrow pn$ has been measured using the CEBAF Large Acceptance Spectrometer (CLAS) at the Thomas Jefferson National Accelerator Facility (JLab) for six photon-energy bins between 1.1 and 2.3 GeV, and proton angles in the center-of-mass frame, $\theta_{c.m.}$, between $25^\circ$ and $160^\circ$. These are the first measurements of beam-spin asymmetries at $\theta_{c.m.}=90^\circ$ for photon-beam energies above 1.6 GeV, and the first measurements for angles other than $\theta_{c.m.}=90^\circ$. The angular and energy dependence of $\Sigma$ is expected to aid in the development of QCD-based models to understand the mechanisms of deuteron photodisintegration in the transition region between hadronic and partonic degrees of freedom, where both effective field theories and perturbative QCD cannot make reliable predictions., Comment: 18 pages

First results from the longitudinally polarized frozen-spin target (FROST) program are reported. The double-polarization observable E, for the reaction $\vec \gamma \vec p \to \pi^+n$, has been measured using a circularly polarized tagged-photon beam, with energies from 0.35 to 2.37 GeV. The final-state pions were detected with the CEBAF Large Acceptance Spectrometer in Hall B at the Thomas Jefferson National Accelerator Facility. These polarization data agree fairly well with previous partial-wave analyses at low photon energies. Over much of the covered energy range, however, significant deviations are observed, particularly in the high-energy region where high-L multipoles contribute. The data have been included in new multipole analyses resulting in updated nucleon resonance parameters. We report updated fits from the Bonn-Gatchina, J\"ulich, and SAID groups., Comment: 6 pages, 3 figures