An Autoencoder-Based Response Time Model and Its Application in Anomaly Detection

Most psychometric models of response times are primarily theory-driven, meaning they are based on various sets of assumptions about how the data should behave. Although useful in certain contexts, such models are often inadequate for the complexities of realistic testing situations and display a poor fit on empirical data. Therefore, as a functional alternative, the present study proposes a data-driven approach, an autoencoder-based response time model, to modeling response times of correctly answered responses. Also, this study introduces the application of the proposed model in anomaly detection (including aberrant examinee and item detection). The result shows this model has an acceptable performance in both response time modeling and anomaly detection.

Mitigating Cold Start Problem in Serverless Computing with Function Fusion

[sangyoon]As Artificial Intelligence (AI) is becoming ubiquitous in many applications, serverless computing is also emerging as a building block for developing cloud-based AI services. Serverless computing has received much interest because of its simplicity, scalability, and resource efficiency. However, due to the trade-off with resource efficiency, serverless computing suffers from the cold start problem, that is, a latency between a request arrival and function execution[sangyoon] that is encountered due to resource provisioning. [sangyoon]In serverless computing, functions can be composed as workflows to process a complex task, and the cold start problem has a significant influence on workflow response time because the cold start can occur in each function.The cold start problem significantly influences the overall response time of workflow that consists of functions because the cold start may occur in every function within the workflow. Function fusion can be one of the solutions to mitigate the cold start latency of a workflow. If two functions are fused into a single function, the cold start of the second function is removed; however, if parallel functions are fused, the workflow response time can be increased because the parallel functions run sequentially even if the cold start latency is reduced. This study presents an approach to mitigate the cold start latency of a workflow using function fusion while considering a parallel run. First, we identify three latencies that affect response time, present a workflow response time model considering the latency, and efficiently find a fusion solution that can optimize the response time on the cold start. Our method shows a response time of 28–86% of the response time of the original workflow in five workflows.

Understanding Speed–Accuracy Processing Dynamics in Aphasia Using Response Time Modeling

People with aphasia demonstrate language impairments evident in both performance accuracy and processing speed, but the direct relationship between accuracy and speed requires further consideration. This article describes two recent attempts to make quantitative progress in this domain using response time modeling: the diffusion model (Ratcliff, 1978) applied to two-choice tasks and a multinomial ex-Gaussian model applied to picture naming. The diffusion model may be used to characterize core linguistic processing efficiency and speed–accuracy tradeoffs independently, and research suggests that maladaptive speed–accuracy tradeoffs lead to performance impairments in at least some people with aphasia. The multinomial ex-Gaussian response time model of picture naming provides a simple and straightforward way to estimate the optimal response time cutoffs for individual people with aphasia (i.e., the cutoff where additional time is unlikely to lead to a correct response). While response time modeling applied to aphasia research is at an early stage of development, both the diffusion model and multinomial ex-Gaussian response time model of picture naming show promise and should be further developed in future work. This article also provides preliminary recommendations for clinicians regarding how to conceptualize, identify, and potentially address maladaptive speed–accuracy tradeoffs for people with aphasia.

Optimising Subsea Well Source Control Timeframes

ABSTRACT In response to the Montara and Macondo subsea well incidents in 2009–10, the industry's knowledge of and ability to respond to a subsea source control (SSC) event has greatly improved. Industry has invested heavily in its response capabilities and established best practices to resolve future incidents that may arise in the offshore oil and gas operations. The investment has driven rapid advancements in science, engineering, and new technological equipment developments to establish a higher standard for SSC preparedness and readiness. The industry now has a high confidence in its ability to deal with a subsea well release. The growth in capability has led to many variations in equipment and response plans, which has led to complexity in an already highly technical field. To reduce the complexity, common understanding is required of all the actions that comprise a SSC response, the linkages and dependencies between all the actions, and the critical path items that influence the overall timeframes of regaining control of the well. With a common understanding of the response plan comes enhanced industry, regulator and community confidence in the ability of the oil and gas industry to appropriately manage its environmental and social impacts. To help with this effort, the International Association of Oil and Gas Producers (IOGP) has produced reports 592, 594 and 595. Report 594 is a guideline that can be used to support subsea source control response planning and Report 595 addresses capping stack design and operational reliability. IOGP Report 592 - Subsea Capping Response Time Model Toolkit User Guide, was completed in December 2019. It was jointly developed by IOGP and the Australian National Offshore Petroleum Safety and Environmental Authority (NOPSEMA). This report involved the creation of a digital subsea response time model that is freely available with a number of different software templates. The objective was to create a common standardized document that described the processes for preparing and implementing a subsea well blowout response in a timeline format, and in doing so, identify and communicate critical path activities, areas that can be prioritised pre-response, be easily transferrable to other parties to support mutual aid activities and, should the need arise, be used as an actual response project planning tool. This paper informs readers of these resources and explains the reasoning behind their creation.

A MECS redistribution algorithm for SDN-enable MEC using response time and transmission overhead

The software-defined networks-enable mobile edge computing (SDN-enable MEC) architecture, which integrates SDN and MEC technologies, realizes the flexibility and dynamic management of the underlying network resources by the MEC, reduces the distance between the access terminal and computing resources and network resources, and increases the terminal's access to resources. However, the static distribution relationship between MEC servers (MECSs) and controllers in the multi-controller architecture may result in unbalanced load distribution among the controllers, which would degrade network performance. In this paper, a multi-objective optimization MECS redistribution algorithm (MOSRA) is proposed to decrease the response time and overhead. A controller response time model and link transmit overhead model are introduced as basis of an evolutionary algorithm which is proposed to optimize MECS redistribution. The proposed algorithm aims to select an available sub-optimizes individual by using a strategy based coordination transformation from Pareto Front. That is, when the master controller of the MECS is redistributed, both of the network overhead of the MECS to the controller and the response time of the controller to the MECS processing request are optimized. Finally, the simulation results demonstrate that the MOSRA can solve the redistribution problem in different network load levels and different network sizes within the effective time, and has a lower control plane response time, while making the edge network plane transmission overhead lower, compared with other algorithms .

A Deterministic Gated Lognormal Response Time Model to Identify Examinees with Item Preknowledge

Response time (RT) information has recently attracted a significant amount of attention in the literature as it may provide meaningful information about item preknowledge. In this study, a Deterministic Gated Lognormal Response Time (DG-LNRT) model is proposed to identify examinees with potential item preknowledge using RT information. The proposed model is applied to a real experimental dataset provided by Toton and Maynes (2019) in which item preknowledge was manipulated, and its performance is demonstrated. Then, the performance of the DG-LNRT model is investigated through a simulation study. The model is estimated using the Bayesian framework via Stan. The results indicate that the proposed model is viable and has the potential to be useful in detecting cheating by using response time differences between compromised and uncompromised items.

Two Parameter Logistics Model with Lognormal Response Time for Computer-based Testing

In addition to the information on response pattern/accuracy, Computer-Based Testing (CBT) can also generate information on response time. This research contributes to develop two-parameter logistics model with random variables of response time for CBT power test. Based on that model, the probability of the test taker answering the test questions correctly is influenced by the test taker’s ability, the test question’s discriminating power, the question’s level of difficulty, the delay due to the test question’s factors and response time. The development of this model aims to improve the parameter estimation of the logistics model on Item Response Theory (IRT) which does not consider the response time of the model. This model is simultaneously developed using joint distribution concept, by multiplying the conditional distribution of response accuracy (two-parameter logistics model) by response time with marginal distribution of response time. The marginal distribution chosen in this study is lognormal distribution because it has positive value in the form of positive skewed according to the characteristics of the response time. To prove the model is suitable for power test, is tested using CBT data. The study found that the simultaneous model generated from the multiplication between the two-parameter logistics model integrated with response time and the lognormal response time model is an appropriate model for CBT power test.