Обнаружение транспортных средств на изображениях загородных шоссе на основе метода Single shot multibox Detector

Hawaii International Conference on System Sciences

To implement a program functionality, developers can reuse previously written code snippets by searching through a large-scale codebase. Over the years, many code search tools have been proposed to help developers. The existing approaches often treat source code as textual documents and utilize information retrieval models to retrieve relevant code snippets that match a given query. These approaches mainly rely on the textual similarity between source code and natural language query. They lack a deep understanding of the semantics of queries and source code. In this paper, we propose a novel deep neural network named CODEnn (Code-Description Embedding Neural Network). Instead of matching text similarity, CODEnn jointly embeds code snippets and natural language descriptions into a high-dimensional vector space, in such a way that code snippet and its corresponding description have similar vectors. Using the unified vector representation, code snippets related to a natural language query can be retrieved according to their vectors. Semantically related words can also be recognized and irrelevant/noisy keywords in queries can be handled. As a proof-of-concept application, we implement a code search tool named DeepCS using the proposed CODEnn model. We empirically evaluate DeepCS on a large scale codebase collected from GitHub. The experimental results show that our approach can effectively retrieve relevant code snippets and outperforms previous techniques.

Deep code search

Background: New possibilities for mHealth have arisen by means of the latest advances in mobile communications and technologies. With more than 1 billion smartphones and 100 million tablets around the world, these devices can be a valuable tool in health care management. Every aid for health care is welcome and necessary as shown by the more than 50 million estimated deaths caused by illnesses or health conditions in 2008. Some of these conditions have additional importance depending on their prevalence. Objective: To study the existing applications for mobile devices exclusively dedicated to the eight most prevalent health conditions by the latest update (2004) of the Global Burden of Disease (GBD) of the World Health Organization (WHO): iron-deficiency anemia, hearing loss, migraine, low vision, asthma, diabetes mellitus, osteoarthritis (OA), and unipolar depressive disorders. Methods: Two reviews have been carried out. The first one is a review of mobile applications in published articles retrieved from the following systems: IEEE Xplore, Scopus, ScienceDirect, Web of Knowledge, and PubMed. The second review is carried out by searching the most important commercial app stores: Google play, iTunes, BlackBerry World, Windows Phone Apps+Games, and Nokia's Ovi store. Finally, two applications for each condition, one for each review, were selected for an in-depth analysis. Results: Search queries up to April 2013 located 247 papers and more than 3673 apps related to the most prevalent conditions. The conditions in descending order by the number of applications found in literature are diabetes, asthma, depression, hearing loss, low vision, OA, anemia, and migraine. However when ordered by the number of commercial apps found, the list is diabetes, depression, migraine, asthma, low vision, hearing loss, OA, and anemia. Excluding OA from the former list, the four most prevalent conditions have fewer apps and research than the final four. Several results are extracted from the in-depth analysis: most of the apps are designed for monitoring, assisting, or informing about the condition. Typically an Internet connection is not required, and most of the apps are aimed for the general public and for nonclinical use. The preferred type of data visualization is text followed by charts and pictures. Assistive and monitoring apps are shown to be frequently used, whereas informative and educational apps are only occasionally used. Conclusions: Distribution of work on mobile applications is not equal for the eight most prevalent conditions. Whereas some conditions such as diabetes and depression have an overwhelming number of apps and research, there is a lack of apps related to other conditions, such as anemia, hearing loss, or low vision, which must be filled. [J Med Internet Res 2013;15(6):e120]

Mobile Health Applications for the Most Prevalent Conditions by the World Health Organization: Review and Analysis

Fault tolerance is the study of reliable computation using unreliable components. With a given noise model, can one still reliably compute? For example, one can run many copies of a classical calculation in parallel, periodically using majority gates to catch and correct faults. Von Neumann showed in 1956 that if each gate fails independently with probability p, flipping its output bit 0 $ 1, then such a fault tolerance scheme still allows for arbitrarily reliable computation provided that p is below some constant threshold (whose value depends on the model details) [10]. In a quantum computer, the basic gates are much more vulnerable to noise than classical transistors – after all, depending on the implementation, they are manipulating single electron spins, photon polarizations, and similarly fragile subatomic particles. It might not be possible to engineer systems with noise rates less than 10 , or perhaps 10 , per gate. Additionally, the phenomenon of entanglement makes quantum systems inherently fragile. For example, in Schrödinger’s cat state – an equal superposition between a living cat and a dead cat, often idealized as 1= p 2.j0ni C j1ni/ – an interaction with just one quantum bit (“qubit”) can collapse, or decohere, the entire system. Fault tolerance techniques will therefore be essential for achieving the considerable potential of quantum computers. Practical fault tolerance techniques will need to control high noise rates and do so with low overhead, since qubits are expensive. Quantum systems are continuous, not discrete, so there are many possible noise models. However, the essential features of quantum noise for fault tolerance results can be captured by a simple discrete model similar to the one Von Neumann used. The main difference is that, in addition to bit-flip X errors which swap 0 and 1, there can also be phase-flip Z errors which swap jCi 1=p2.j0i C j1i/ and j i 1=p2.j0i j1i/ (Fig. 1). A noisy gate is modeled as a perfect gate followed by independent introduction of X, Z, or Y (which is both X and Z) errors with respective probabilities pX; pZ; pY. One popular model is independent depolarizing noise .pX D pZ D pY p=3/; a depolarized qubit is completely randomized. Faulty measurements and preparations of single-qubit states must additionally be modeled, and there can be memory noise on resting qubits. It is often assumed that measurement results can

Fault-Tolerant Quantum Computation.

Status and trends of mobile-health applications for iOS devices: A developer's perspective

When a new bug report is received, developers usually need to reproduce the bug and perform code reviews to find the cause, a process that can be tedious and time consuming. A tool for ranking all the source files of a project with respect to how likely they are to contain the cause of the bug would enable developers to narrow down their search and potentially could lead to a substantial increase in productivity. This paper introduces an adaptive ranking approach that leverages domain knowledge through functional decompositions of source code files into methods, API descriptions of library components used in the code, the bug-fixing history, and the code change history. Given a bug report, the ranking score of each source file is computed as a weighted combination of an array of features encoding domain knowledge, where the weights are trained automatically on previously solved bug reports using a learning-to-rank technique. We evaluated our system on six large scale open source Java projects, using the before-fix version of the project for every bug report. The experimental results show that the newly introduced learning-to-rank approach significantly outperforms two recent state-of-the-art methods in recommending relevant files for bug reports. In particular, our method makes correct recommendations within the top 10 ranked source files for over 70% of the bug reports in the Eclipse Platform and Tomcat projects.

Learning to rank relevant files for bug reports using domain knowledge

The application of information retrieval techniques to search tasks in software engineering is made difficult by the lexical gap between search queries, usually expressed in natural language (eg English), and retrieved documents, usually expressed in code (eg programming languages) This is often the case in bug and feature location, community question answering, or more generally the communication between technical personnel and non-technical stake holders in a software project In this paper, we propose bridging the lexical gap by projecting natural language statements and code snippets as meaning vectors in a shared representation space In the proposed architecture, word embeddings are first trained on API documents, tutorials, and reference documents, and then aggregated in order to estimate semantic similarities between documents Empirical evaluations show that the learned vector space embeddings lead to improvements in a previously explored bug localization task and a newly defined task of linking API documents to computer programming questions

From word embeddings to document similarities for improved information retrieval in software engineering

Three-dimensional online virtual worlds such as second life support avatar-based communications, a wide spectrum of online activities, and development of various in-world teaching and learning tools. We have experimented with second life in two computer science classes, one at Ohio University, the other at the University of Mary Washington, to enhance software engineering education. We used Second Life as an innovative collaboration and communication tool both in and outside classroom to help facilitate teamwork and interactions among student project team members. Second Life was also used as the virtual office for instructors and teaching assistants to answer students' questions during office hours. In addition, we developed two multi-player online software engineering educational games in second life, one based on the Groupthink software specification exercise developed at M.I.T., and the other based on the SimSE game (a 2-D single player game) developed at UC Irvine. By playing these two games, students learned fundamentals of software specification activities and principles of software development processes. In the paper, we will share our experience of using second life in two software engineering classes, and discuss its pros and cons based on the data collected from student surveys.

Enhancing software engineering education using teaching aids in 3-D online virtual worlds

When a new bug report is received, developers usually need to reproduce the bug and perform code reviews to find the cause, a process that can be tedious and time consuming. A tool for ranking all the source files with respect to how likely they are to contain the cause of the bug would enable developers to narrow down their search and improve productivity. This paper introduces an adaptive ranking approach that leverages project knowledge through functional decomposition of source code, API descriptions of library components, the bug-fixing history, the code change history, and the file dependency graph. Given a bug report, the ranking score of each source file is computed as a weighted combination of an array of features, where the weights are trained automatically on previously solved bug reports using a learning-to-rank technique. We evaluate the ranking system on six large scale open source Java projects, using the before-fix version of the project for every bug report. The experimental results show that the learning-to-rank approach outperforms three recent state-of-the-art methods. In particular, our method makes correct recommendations within the top 10 ranked source files for over 70 percent of the bug reports in the Eclipse Platform and Tomcat projects.

Chang Liu

Papers

Status and trends of mobile-health applications for iOS devices: A developer's perspective

Learning to rank relevant files for bug reports using domain knowledge

From word embeddings to document similarities for improved information retrieval in software engineering

Enhancing software engineering education using teaching aids in 3-D online virtual worlds

Mapping Bug Reports to Relevant Files: A Ranking Model, a Fine-Grained Benchmark, and Feature Evaluation