The maximum cut problem finds a partition of a graph that maximizes the number of crossing edges. When the graph is dense or is sampled based on certain planted assumptions, there exist polynomial-time approximation s...
详细信息
ISBN:
(数字)9781665485104
ISBN:
(纸本)9781665485104
The maximum cut problem finds a partition of a graph that maximizes the number of crossing edges. When the graph is dense or is sampled based on certain planted assumptions, there exist polynomial-time approximation schemes that given a fixed epsilon > 0, find a solution whose value is at least 1 - epsilon of the optimal value. This paper presents another random model relating to both successful cases. Consider an n-vertex graph G whose edges are sampled from an unknown dense graph H independently with probability p = Omega(1/root log n);this input graph G has O(n(2)/root log n) edges and is no longer dense. We show how to modify a PTAS by de la Vega for dense graphs to find an (1 - epsilon)-approximate solution for G. Although our algorithm works for a very narrow range of sampling probability p, the sampling model itself generalizes the planted models fairly well.
In softwareengineering, behavioral state machine models play a crucial role in validating system behavior and maintaining correctness. This paper proposes an extension of an existing architecture for automatically le...
详细信息
ISBN:
(纸本)9798350376975;9798350376968
In softwareengineering, behavioral state machine models play a crucial role in validating system behavior and maintaining correctness. This paper proposes an extension of an existing architecture for automatically learning state machine models of client-server systems that automates processes such as regression detection and test case generation, and guides the development of new features. The learned models help identify potential implementation issues of clients, servers, their interactions, as well as the protocols themselves. The architecture also enhances the debugging process and ensures comprehensive system coverage. By employing the LTSDiff algorithm, the method efficiently detects behavioral changes due to software updates, preventing unintended consequences on system performance. Consequently, the automatically generated state machine models can be used as evidence in security, safety, and reliability assurance, providing a valuable tool for development, testing, and maintenance of complex software systems. The learned state machines and detected changes correctly model the behavior of a client-server system to a specified depth at the level of an active outside adversary with the capability to read, replay, replace, or block any message.
The transformation of pseudocode to Python is vital as it enables students to concentrate on the algorithms while not being distracted by the syntax and also is the key stage in software development and computer scien...
详细信息
Graduates distinguish themselves by being creative, innovative, and showing leadership, but such instructional topics are usually formally emphasized in business curricula. computerscience students could minor in ent...
详细信息
ISBN:
(纸本)9798400701382
Graduates distinguish themselves by being creative, innovative, and showing leadership, but such instructional topics are usually formally emphasized in business curricula. computerscience students could minor in entrepreneurship or other business-related fields in order to foster their innovative mindset and obtain a business acumen that complements the technical expertise they obtain in their majors, but such minors usually require at least five additional courses. We discuss the formal incorporation of entrepreneurship in the computerscience curriculum as part of a clinic model that includes real-world experiences, at the expense of two 1-credit hands-on clinics that can be embedded into the number of credits required for the computerscience degree. After experimenting with various models and activities over a period of several years, we present our current iteration that was successfully deployed for the past four years as a course framework that combines both entrepreneurial and technical aspects into a two-semester softwareengineering course sequence with assigned clinic experience during the junior year. Students learn how to find and evaluate ideas, build rapid prototypes, test hypotheses, use different types of business models and financial analysis, market their software products, and understand how to start a business. Students are able to pitch and refine their ideas with various audiences and compete in entrepreneurship competitions. From dreaming ideas to creating and managing teams, computerscience students are guided through the process by faculty, entrepreneurs, and potential investors. We present the steps and activities that complement our framework, with case studies and lesson learned.
In the analysis of drone aerial images, object detection tasks are particularly challenging, especially in the presence of complex terrain structures, extreme differences in target sizes, suboptimal shooting angles, a...
详细信息
In the analysis of drone aerial images, object detection tasks are particularly challenging, especially in the presence of complex terrain structures, extreme differences in target sizes, suboptimal shooting angles, and varying lighting conditions, all of which exacerbate the difficulty of recognition. In recent years, the DETR model based on the Transformer architecture has eliminated traditional post-processing steps such as NMS(Non-Maximum Suppression), thereby simplifying the object detection process and improving detection accuracy, which has garnered widespread attention in the academic community. However, DETR has limitations such as slow training convergence, difficulty in query optimization, and high computational costs, which hinder its application in practical fields. To address these issues, this paper proposes a new object detection model called OptiDETR. This model first employs a more efficient hybrid encoder to replace the traditional Transformer encoder. The new encoder significantly enhances feature processing capabilities through internal and cross-scale feature interaction and fusion logic. Secondly, an IoU (Intersection over Union) aware query selection mechanism is introduced. This mechanism adds IoU constraints during the training phase to provide higher-quality initial object queries for the decoder, significantly improving the decoding performance. Additionally, the OptiDETR model integrates SW-Block into the DETR decoder, leveraging the advantages of Swin Transformer in global context modeling and feature representation to further enhance the performance and efficiency of object detection. To tackle the problem of small object detection, this study innovatively employs the SAHI algorithm for data augmentation. Through a series of experiments, It achieved a significant performance improvement of more than two percentage points in the mAP (mean Average Precision) metric compared to current mainstream object detection models. Furthermore, ther
Today, there are many critical data sources generating multi-seasonal time series data that demand efficient and accurate anomaly detection. Current existing seasonal forecasting models struggle with multiple seasonal...
详细信息
This softwareengineering (SE) is the core research area of the software industry, and numerous algorithms and frameworks are proposed every day. softwareengineering is the discipline that keeps up with recent develo...
详细信息
Applying standard softwareengineering practices to neural networks is challenging due to the lack of high-level abstractions describing a neural network's behavior. To address this challenge, we propose to extrac...
详细信息
ISBN:
(数字)9783031308260
ISBN:
(纸本)9783031308253;9783031308260
Applying standard softwareengineering practices to neural networks is challenging due to the lack of high-level abstractions describing a neural network's behavior. To address this challenge, we propose to extract high-level task-specific features from the neural network internal representation, based on monitoring the neural network activations. The extracted feature representations can serve as a link to high-level requirements and can be leveraged to enable fundamental softwareengineering activities, such as automated testing, debugging, requirements analysis, and formal verification, leading to better engineering of neural networks. Using two case studies, we present initial empirical evidence demonstrating the feasibility of our ideas.
Log files generated by software systems can be utilized as a valuable resource in data-driven approaches to improve the system health and stability. These files often contain valuable information about runtime executi...
详细信息
ISBN:
(纸本)9798350381566;9798350381559
Log files generated by software systems can be utilized as a valuable resource in data-driven approaches to improve the system health and stability. These files often contain valuable information about runtime execution, and their effective monitoring requires analyzing an increasingly large volume of data logs. In this paper, a graph mining technique for log parsing is presented, which is source agnostic to the system. This means that the technique can function regardless of the source of the logs, making it more scalable and reusable. Unlike the existing approaches that rely heavily on domain knowledge and regular expression patterns, the proposed approach uses graph models and semantic analysis to detect data patterns with minimal user input. This makes it easy to implement it in a variety of scenarios where application-based logs may differ significantly. The proposed parsing technique is evaluated over seven datasets. It achieves the best performance on the Thunderbird dataset, where the technique takes 3.87 seconds for 2000 logs, while obtaining precision, recall and F1 measure higher than 0.99.
Cloud storage is crucial for managing large datasets, but dependence on a single cloud space raises security concerns. Conversely, Distributed Ledger Technology (DLT) provides a secure cloud-based storage system opera...
详细信息
暂无评论