The source code suggestions provided by current integrated development environment (IDEs) are mostly dependent on static type learning. These suggestions often end up proposing irrelevant suggestions for a peculiar co...
详细信息
The source code suggestions provided by current integrated development environment (IDEs) are mostly dependent on static type learning. These suggestions often end up proposing irrelevant suggestions for a peculiar context. Recently, deep learning-based approaches have shown great potential in the modelling of source code for various software engineering tasks. However, these techniques lack adequate generalisation and resistance to acclimate the use of such models in a real-world software development environment. This Letter presents DeepVS, an end-to-end deep neural code completion tool that learns from existing codebases by exploiting the bidirectional gated recurrent unit (BiGRU) neural net. The proposed tool is capable of providing source code suggestions instantly in an IDE by using pre-trained BiGRU neural net. The evaluation of this work is two-fold, quantitative and qualitative. Through extensive evaluation on ten real-world open-source software systems, the proposed method shows significant performance enhancement and its practicality. Moreover, the results also suggest that DeepVS tool is capable of suggesting zero-day (unseen) code tokens by learning coding patterns from real-world software systems.
This thesis describes the design and use of an online system of computerized lessons built to teach reading and writing to students with learning disabilities as part of a large research project on specific learning d...
详细信息
This thesis describes the design and use of an online system of computerized lessons built to teach reading and writing to students with learning disabilities as part of a large research project on specific learning disabilities in written language. The system comprises two main components: HAWK, the portion that teaches reading and writing skills, and Kokopelli’s World, the addition to HAWK that introduces programming instruction. Kokopelli’s World is a novel blocks-based coding environment designed to cater to the needs of dyslexic students. The thesis also presents two user studies that explore the relationship between programming and dyslexia. Results suggest that different measures of dyslexia may have different correlation with programming success. The thesis also presents detailed coding experiences of two dyslexic students and discusses them in terms of a newly constructed set of recurring programming patterns. These studies have identified a set of hypotheses about how dyslexic students relate to computer programming that can help structure future research on programming by dyslexic children.
Recently, traditional tunable coding metasurfaces integrated with active components or materials have exhibited not only many functionalities, but also reconfigurable responses, which can execute dynamic beam control....
详细信息
Recently, traditional tunable coding metasurfaces integrated with active components or materials have exhibited not only many functionalities, but also reconfigurable responses, which can execute dynamic beam control. However, these metasurfaces are difficult to realize effective and active terahertz wavefront manipulation due to the inherent defects of the active components or materials, such as high loss, polarization dependence, slow response, and so on, dramatically hindering the potential practical applications of terahertz technologies and devices. To implement programmable manipulation of terahertz beams, here, we propose the reflective-type hybrid graphene-metal coding metasurfaces whose meta-atoms are composed of the different C-shaped metallic split-ring resonators integrated with the graphene, thus, obtaining different coding states by rotating structure and doping graphene. Based on these merits, a couple of coding metasurfaces can effectively achieve various functionalities, including single-and multi-beam deflections, diffusive scattering, and vortex beam, via changed coding patterns. Moreover, these functionalities can further realize dynamic reconfiguration by tuning Fermi energy of graphene. Therefore, the proposed coding metasurfaces would show great potential application prospects in terahertz sensing, display, telecommunication, and imaging.
In recent years, Coded Aperture Snapshot Spectroscopic Imaging System (CASSI) has attracted more and more attention. However, in the case of undersampling, the effect of the reconstructed image is often un-satisfactor...
详细信息
In recent years, Coded Aperture Snapshot Spectroscopic Imaging System (CASSI) has attracted more and more attention. However, in the case of undersampling, the effect of the reconstructed image is often un-satisfactory. For spectral imaging system, the slower the imaging speed, the higher the image resolution of hyperspectral images, and vice versa. To solve this problem, in this paper, we design an undersam-pling rapid coded aperture spectrometer system. Based on the analysis of the internal relationship be-tween coded spectral imaging and single pixel imaging, the order of the discrete cosine transform (DCT) coding patterns is selected by using the energy concentration property of the DCT and the rich spatial information contained in the RGB images. In addition, we built a spectral image enhancement network to improve the quality of the reconstructed images. At the sampling rate of 0.0625, the PSNR, SSIM and SAM of the spectral images reconstructed based on our system are 43.8341, 0.9869 and 2.0862, respec-tively, the reconstruction effect of this method is significantly superior to other methods, and this method provides an idea to address the contradiction between image resolution and imaging speed to a certain extent, so as to achieve high speed and high resolution hyperspectral imaging. (c) 2023 Elsevier B.V. All rights reserved.
It is not secret that generative AI, especially in the form of large language models (LLMs), is extremely popular today. One might go so far as to say that it's eaten the world. It may be a bubble, or it may last-...
详细信息
It is not secret that generative AI, especially in the form of large language models (LLMs), is extremely popular today. One might go so far as to say that it's eaten the world. It may be a bubble, or it may last-though the death of cryptocurrencies has long been predicted, as I write this Bitcoin has just reached an all-time high value against the American dollar-but for now and at least the next few years, generative AI will be with us. As people who care about security and privacy, we need to understand the implications of it: is it good or bad for our field, and if the latter, what should we do about it? Ignoring it is not an option.
暂无评论