In this paper, we first discuss the challenges of adapting an already trained DNN-based anomaly detector with knowledge mined during the execution of the main system. Then, we present a framework for the continual lea...
详细信息
ISBN:
(数字)9781728177359
ISBN:
(纸本)9781728177366
In this paper, we first discuss the challenges of adapting an already trained DNN-based anomaly detector with knowledge mined during the execution of the main system. Then, we present a framework for the continual learning of anomaly detectors, which records in-field behavioural data to determine what data are appropriate for adaptation. We evaluated our framework to improve an anomaly detector taken from the literature, in the context of misbehavior prediction for self-driving cars. Our results show that our solution can reduce the false positive rate by a large margin and adapt to nominal behaviour changes while maintaining the original anomaly detection capability.
This paper overviews the solution of the Schrödinger equation for the case of one-dimensional infinite potential well with a neural network approach. Using a single hidden layer neural network, which is proved to...
ISBN:
(数字)9789532900996
ISBN:
(纸本)9781728175386
This paper overviews the solution of the Schrödinger equation for the case of one-dimensional infinite potential well with a neural network approach. Using a single hidden layer neural network, which is proved to be a universal function approximator, and by exploiting the automatic differentiation capabilities, it is possible to achieve accurate values of the wave function and eigenvalues of the ground state. The loss function with integrated physical knowledge is set up as an unconstrained nonlinear problem and parameters of a neural network are being learnt in a completely unsupervised manner. Such a technique could potentially serve as a door opener for solving high-dimensional quantum mechanics problems, otherwise tedious to set up for standard mesh-based numerical methods.
Recently, ontologies and semantic data technologies have increasingly come back into the focus of research due to the emerging use of knowledge graphs. However, even though the modeling and simulation community has re...
ISBN:
(数字)9781728189567
ISBN:
(纸本)9781728189574
Recently, ontologies and semantic data technologies have increasingly come back into the focus of research due to the emerging use of knowledge graphs. However, even though the modeling and simulation community has recognized the potential of using this technology for the modeling process, for example for automatic model generation, adaptation or to represent simulation expert knowledge, a general and reusable approach for the aforementioned purposes is still missing. Therefore, in this paper a state of the art review for using knowledge representation during modeling and simulation processes of complex technical systems is conducted, such as factories or process plants with specific focus on the production and logistic domain. based on that, requirements and benefits of knowledge graphs in this specific domain are evaluated.
Testing of web APIs is nowadays more critical than ever before, as they are the current standard for software integration. A bug in an organization's web API could have a huge impact both internally (services rely...
详细信息
ISBN:
(数字)9781450371223
ISBN:
(纸本)9781728165288
Testing of web APIs is nowadays more critical than ever before, as they are the current standard for software integration. A bug in an organization's web API could have a huge impact both internally (services relying on that API) and externally (third-party applications and end users). Most existing tools and testing approaches require writing tests or instrumenting the system under test (SUT). The main aim of this dissertation is to take web API testing to an unprecedented level of automation and thoroughness. To this end, we plan to apply artificial intelligence (AI) techniques for the autonomous detection of software failures. Specifically, the idea is to develop intelligent programs (we call them "bots") capable of generating hundreds, thousands or even millions of test inputs and to evaluate whether the test outputs are correct based on: 1) patterns learned from previous executions of the SUT; and 2) knowledge gained from analyzing thousands of similar programs. Evaluation results of our initial prototype are promising, with bugs being automatically detected in some real-world APIs.
The proceedings contain 34 papers. The topics discussed include: is virtual reality product development different? an empirical study on VR product development practices;ThrustHetero: a framework to simplify heterogen...
ISBN:
(纸本)9781450362153
The proceedings contain 34 papers. The topics discussed include: is virtual reality product development different? an empirical study on VR product development practices;ThrustHetero: a framework to simplify heterogeneous computing platform programming using design abstraction;an approach to identify use case scenarios from textual requirements specification;change-proneness of object-oriented software using combination of feature selection techniques and ensemble learning techniques;enhancing test cases generated by concolic testing;a knowledge centric approach to conceptualizing robotic solutions;making sense of actor behavior: an algebraic filmstrip pattern and its implementation;a software framework for adaptive signal analytics based on autonomic service components;modeling and coverage analysis of programs with exception handling;the value of software architecture recovery for maintenance;evolution traceability roadmap for business processes;key factors in scaling up agile team in matrix organization;and a case study exploring supply chain systems using actor based simulation.
Community detection partitions users in social networks into sub-groups according to structural or behavioral similarities, which had been widely adopted by a lot of applications such as friend recommendation, precisi...
详细信息
ISBN:
(纸本)9783030295516;9783030295509
Community detection partitions users in social networks into sub-groups according to structural or behavioral similarities, which had been widely adopted by a lot of applications such as friend recommendation, precision marketing, etc. In this paper, we propose a location-interest-aware community detection approach for mobile social networks. Specifically, we develop a spatial-temporal topic model to describe users' location interest, and introduce an auto encoder mechanism to represent users' location features and social network features as low-dimensional vectors, based on which a community detection algorithm is applied to divide users into sub-graphs. We conduct extensive experiments based on a real-world mobile social network dataset, which demonstrate that the proposed community detection approach outperforms the baseline algorithms in a variety of performance metrics.
A model for the management of capstone courses of engineering careers is proposed, in particular in Systems engineering, Computer Science, Computer engineering and related. The model is based on Project-based Learning...
详细信息
The rapid development of the Internet and the increase in the magnitude of the dissemination of knowledge and information have made it difficult for people to quickly find knowledge and information that suits them fro...
详细信息
ISBN:
(数字)9781728152448
ISBN:
(纸本)9781728152455
The rapid development of the Internet and the increase in the magnitude of the dissemination of knowledge and information have made it difficult for people to quickly find knowledge and information that suits them from the huge range of knowledge. Traditional knowledge recommendation algorithms often face many deficiencies in their use, and perform poorly in actual applications. This paper proposes a knowledge recommendation algorithm that calculates the correlation through the tacit knowledge relationship. It uses vector quantification of the implicit knowledge relationship, and uses the concept of information entropy to calculate the maximum recommendable range, uses expert samples to learn to assist decision-making, and adjusts the results through variable factors. Experiments show that the algorithm proposed in this paper can effectively improve the accuracy and efficiency of recommendation and ensure the diversity of recommendation results.
Content less than two hundred words like comments or review statements is known as a short text. Short text classification is useful for automatically categorizing sentence into predefined group. There are several tra...
详细信息
ISBN:
(纸本)9781728116518
Content less than two hundred words like comments or review statements is known as a short text. Short text classification is useful for automatically categorizing sentence into predefined group. There are several traditional short text classification methods by using bag-of-words with k nearest neighbors (k-NN), Naive Bayes, Maximum entropy, support vector machines (SVMs), and an algorithm based on statistics and rules. The deep learning method is outperformed other methods on classification of short text with normal size of dataset. Some researches classify requirements into functional and non-functional requirements. There is no research on multiclassification of functional requirements with a small dataset particularly for an accounting field. This paper presents an approach to classify short text for a small dataset into multiple categories of functional requirements on the accounting domain. The proposed approach uses an ontology to construct bag-of-words and uses Naive Bayes to classify for small dataset. The experiment is conducted using four hundred of datasets with 5-folds and 10-folds cross validation. The result shows that the method can correctly classify more than 80%. Additionally, comparisons between the ontology-based Naive Bayes method and other methods are investigated.
The PyPI ecosystem has indexed millions of Python libraries to allow developers to automatically download and install dependencies of their projects based on the specified version constraints. Despite the convenience ...
详细信息
ISBN:
(数字)9781450371216
ISBN:
(纸本)9781728165196
The PyPI ecosystem has indexed millions of Python libraries to allow developers to automatically download and install dependencies of their projects based on the specified version constraints. Despite the convenience brought by automation, version constraints in Python projects can easily conflict, resulting in build failures. We refer to such conflicts as Dependency Conflict (DC) issues. Although DC issues are common in Python projects, developers lack tool support to gain a comprehensive knowledge for diagnosing the root causes of these issues. In this paper, we conducted an empirical study on 235 real-world DC issues. We studied the manifestation patterns and fixing strategies of these issues and found several key factors that can lead to DC issues and their regressions. based on our findings, we designed and implemented Watchman, a technique to continuously monitor dependency conflicts for the PyPI ecosystem. In our evaluation, Watchman analyzed PyPI snapshots between 11 Jul 2019 and 16 Aug 2019, and found 117 potential DC issues. We reported these issues to the developers of the corresponding projects. So far, 63 issues have been confirmed, 38 of which have been quickly fixed by applying our suggested patches.
暂无评论