Workshop on Deep Learning and Software Engineering

该奖项支持一个研讨会，以探索深度学习和软件工程之间的协同作用。我们的目标是加速在研究和实践中使用深度学习的研究，通过深度学习的力量改进软件工程的技术和工具。相反，在许多应用领域出现的基于深度学习的系统，需要新的软件工程方法使其正确、可靠和可理解。研讨会将汇集这两个领域的研究人员和实践者，讨论加速深度学习和软件工程交叉研究所需的研究重点和社区资源。深度学习代表了机器通过自动提取给定计算任务的显著特征从数据中学习模式的方式的根本转变，而不是依赖人类的直觉。深度学习方法的特点是由多个层组成的体系结构，这些层根据可学习的参数集对经过它们的数据进行数学转换。这些计算层和参数形成模型，可以根据模型更新参数来训练特定任务，例如图像分类。S在标记的训练数据集上的性能。考虑到软件存储库中可以作为训练数据的大量数据，深度学习技术已经在软件工程研究的一系列任务中取得了进展，包括自动软件修复、代码建议、缺陷预测、恶意软件检测、特征定位等等。研讨会将回顾研究和实践的现状，并就机遇和挑战向社区提供指导。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

This award supports a workshop to explore synergies between Deep Learning and Software Engineering. The goal is to accelerate research that uses Deep Learning in research and practice to improve techniques and tools for Software Engineering through the power of Deep Learning. Conversely, deep-learning based systems, which are emerging in many application domains, need new Software Engineering approaches render them correct, reliable and comprehensible. The workshop will bring together researchers and practitioners in both fields to discuss research priorities community resources needed to accelerate research in the intersection of Deep Learning and Software Engineering. Deep Learning represents a fundamental shift in the manner by which machines learn patterns from data by automatically extracting salient features for a given computational task, as opposed to relying upon human intuition. Deep Learning approaches are characterized by architectures comprised of several layers that perform mathematical transformations, according to sets of learnable parameters, on data passing through them. These computational layers and parameters form models that can be trained for specific tasks, such as image classification, by updating the parameters according to a model?s performance on a labeled set of training data. Given the immense amount of data in software repositories that can serve as training data, deep learning techniques have ushered in advancements across a range of tasks in software engineering research including automatic software repair, code suggestion, defect prediction, malware detection, feature location, and many others. The workshop will review the state of the research and practice and give guidance to the community about opportunities and challenges.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.