在本文中，提出了一种深入的学习方法，可以在全球导航卫星系统（GNSS）剥夺环境中精确定位轮式车辆。在没有GNSS信号的情况下，可以使用关于从车轮编码器记录的车辆（或其他机器人相似的车轮）速度的信息来通过车辆的线性速度的整合来提供用于车辆的连续定位信息流离失所。然而，来自车轮速度测量的位移估计的特征在于不确定因素，其可以表现为车轮滑动或/和对轮胎尺寸或压力的变化，从潮湿和泥泞的道路驱动器或轮胎佩戴。因此，我们利用深度学习的最近进步提出了车轮内径神经网络（WHONET）来学习校正和准确定位所需的车轮速度测量中的不确定性。首先在若干具有挑战性的驾驶场景中评估所提出的WHONET的性能，例如环形交叉路口，锋利的转弯，硬制动和湿路（漂移）。然后，在长期GNSS中断场景中进一步且广泛地评估WHONET的性能，分别在493km的总距离上的长期GNSS中断场景。获得的实验结果表明，在任何180多个行驶之后，所提出的方法能够准确地定位其原始对应物的定位误差高达93％的车辆。 Whonet的实现可以在https://github.com/onyekpeu/whonet找到。

The widespread of offensive content online, such as hate speech and cyber-bullying, is a global phenomenon. This has sparked interest in the artificial intelligence (AI) and natural language processing (NLP) communities, motivating the development of various systems trained to detect potentially harmful content automatically. These systems require annotated datasets to train the machine learning (ML) models. However, with a few notable exceptions, most datasets on this topic have dealt with English and a few other high-resource languages. As a result, the research in offensive language identification has been limited to these languages. This paper addresses this gap by tackling offensive language identification in Sinhala, a low-resource Indo-Aryan language spoken by over 17 million people in Sri Lanka. We introduce the Sinhala Offensive Language Dataset (SOLD) and present multiple experiments on this dataset. SOLD is a manually annotated dataset containing 10,000 posts from Twitter annotated as offensive and not offensive at both sentence-level and token-level, improving the explainability of the ML models. SOLD is the first large publicly available offensive language dataset compiled for Sinhala. We also introduce SemiSOLD, a larger dataset containing more than 145,000 Sinhala tweets, annotated following a semi-supervised approach.

Artificial Intelligence (AI) and its data-centric branch of machine learning (ML) have greatly evolved over the last few decades. However, as AI is used increasingly in real world use cases, the importance of the interpretability of and accessibility to AI systems have become major research areas. The lack of interpretability of ML based systems is a major hindrance to widespread adoption of these powerful algorithms. This is due to many reasons including ethical and regulatory concerns, which have resulted in poorer adoption of ML in some areas. The recent past has seen a surge in research on interpretable ML. Generally, designing a ML system requires good domain understanding combined with expert knowledge. New techniques are emerging to improve ML accessibility through automated model design. This paper provides a review of the work done to improve interpretability and accessibility of machine learning in the context of global problems while also being relevant to developing countries. We review work under multiple levels of interpretability including scientific and mathematical interpretation, statistical interpretation and partial semantic interpretation. This review includes applications in three areas, namely food processing, agriculture and health.

人们经常利用在线媒体（例如Facebook，reddit）作为表达心理困扰并寻求支持的平台。最先进的NLP技术表现出强大的潜力，可以自动从文本中检测到心理健康问题。研究表明，心理健康问题反映在人类选择中所表明的情绪（例如悲伤）中。因此，我们开发了一种新颖的情绪注释的心理健康语料库（Emoment），由2802个Facebook帖子（14845个句子）组成，该帖子从两个南亚国家（斯里兰卡和印度）提取。三名临床心理学研究生参与了将这些职位注释分为八​​类，包括“精神疾病”（例如抑郁症）和情绪（例如，“悲伤”，“愤怒”）。 Emoment语料库达到了98.3％的“非常好”的跨通道协议（即有两个或更多协议），而Fleiss的Kappa为0.82。我们基于罗伯塔的模型的F1得分为0.76，第一个任务的宏观平均F1得分为0.77（即，从职位预测心理健康状况）和第二任务（即相关帖子与定义的类别的关联程度在我们的分类法中）。

数字虚假信息的传播（又称“假新闻”）可以说是互联网上最重要的威胁之一，它可能造成大规模的个人和社会伤害。虚假新闻攻击的敏感性取决于互联网用户在阅读后是否认为虚假新闻文章/摘要是合法的。在本文中，我们试图通过神经认知方法来深入了解用户对以文本为中心的假新闻攻击的敏感性。我们通过脑电图调查了与假/真实新闻有关的神经基础。我们与人类用户进行实验，以彻底调查用户对假/真实新闻的认知处理和认知处理。我们分析了不同类别新闻文章的假/真实新闻检测任务相关的神经活动。我们的结果表明，在人脑处理假新闻与真实新闻的方式上可能没有统计学意义或自动可推断的差异，而当人们受到（真实/假）新闻与安息状态甚至之间的差异时，会观察到明显的差异一些不同类别的假新闻。这一神经认知发现可能有助于证明用户对假新闻攻击的敏感性，这也从行为分析中得到了证实。换句话说，假新闻文章似乎与行为和神经领域的真实新闻文章几乎没有区别。我们的作品旨在剖析假新闻攻击的基本神经现象，并通过人类生物学的极限解释了用户对这些攻击的敏感性。我们认为，对于研究人员和从业者来说，这可能是一个显着的见解楷模

计算光学成像（COI）系统利用其设置中的光学编码元素（CE）在单个或多个快照中编码高维场景，并使用计算算法对其进行解码。 COI系统的性能很大程度上取决于其主要组件的设计：CE模式和用于执行给定任务的计算方法。常规方法依赖于随机模式或分析设计来设置CE的分布。但是，深神经网络（DNNS）的可用数据和算法功能已在CE数据驱动的设计中开辟了新的地平线，该设计共同考虑了光学编码器和计算解码器。具体而言，通过通过完全可区分的图像形成模型对COI测量进行建模，该模型考虑了基于物理的光及其与CES的相互作用，可以在端到端优化定义CE和计算解码器的参数和计算解码器（e2e）方式。此外，通过在同一框架中仅优化CE，可以从纯光学器件中执行推理任务。这项工作调查了CE数据驱动设计的最新进展，并提供了有关如何参数化不同光学元素以将其包括在E2E框架中的指南。由于E2E框架可以通过更改损耗功能和DNN来处理不同的推理应用程序，因此我们提出低级任务，例如光谱成像重建或高级任务，例如使用基于任务的光学光学体系结构来增强隐私的姿势估计，以维护姿势估算。最后，我们说明了使用全镜DNN以光速执行的分类和3D对象识别应用程序。

到2035年，美国电力部门的转型正在进行中，以实现100％无碳污染的电力，以实现这一目标，同时保持安全可靠的电网，需要新的操作范式，以快速准确的决策来制定新的操作范式在动态和不确定的环境中。我们为动态网格重新配置（PHML-DYR）的决策提出了一个新颖的物理知识的机器学习框架，这是电源系统中的关键任务。动态重新配置（DYR）是一个动态设置开关状态的过程，从而导致最佳网格拓扑，从而最大程度地减少线路损耗。为了解决由于决策变量的混合性质而导致的NP硬度的潜在计算复杂性，我们建议使用物理信息信息的ML（PHML），该物理信息（PHML）将操作约束以及拓扑结构和连接性约束集成到神经网络框架中。我们的PHML方法学会同时优化网格拓扑和发电机调度，以满足负载，提高效率并保持在安全的操作范围内。我们证明了PHML-DYR在规范网格上的有效性，显示电力损耗的减少23％，并改善了电压曲线。我们还显示了使用PHML-DYR的数量级以及训练时间的约束违规行为的减少。

In recent hyperspectral unmixing (HU) literature, the application of deep learning (DL) has become more prominent, especially with the autoencoder (AE) architecture. We propose a split architecture and use a pseudo-ground truth for abundances to guide the `unmixing network' (UN) optimization. Preceding the UN, an `approximation network' (AN) is proposed, which will improve the association between the centre pixel and its neighbourhood. Hence, it will accentuate spatial correlation in the abundances as its output is the input to the UN and the reference for the `mixing network' (MN). In the Guided Encoder-Decoder Architecture for Hyperspectral Unmixing with Spatial Smoothness (GAUSS), we proposed using one-hot encoded abundances as the pseudo-ground truth to guide the UN; computed using the k-means algorithm to exclude the use of prior HU methods. Furthermore, we release the single-layer constraint on MN by introducing the UN generated abundances in contrast to the standard AE for HU. Secondly, we experimented with two modifications on the pre-trained network using the GAUSS method. In GAUSS$_\textit{blind}$, we have concatenated the UN and the MN to back-propagate the reconstruction error gradients to the encoder. Then, in the GAUSS$_\textit{prime}$, abundance results of a signal processing (SP) method with reliable abundance results were used as the pseudo-ground truth with the GAUSS architecture. According to quantitative and graphical results for four experimental datasets, the three architectures either transcended or equated the performance of existing HU algorithms from both DL and SP domains.

Ever since the first microscope by Zacharias Janssen in the late 16th century, scientists have been inventing new types of microscopes for various tasks. Inventing a novel architecture demands years, if not decades, worth of scientific experience and creativity. In this work, we introduce Differentiable Microscopy ($\partial\mu$), a deep learning-based design paradigm, to aid scientists design new interpretable microscope architectures. Differentiable microscopy first models a common physics-based optical system however with trainable optical elements at key locations on the optical path. Using pre-acquired data, we then train the model end-to-end for a task of interest. The learnt design proposal can then be simplified by interpreting the learnt optical elements. As a first demonstration, based on the optical 4-$f$ system, we present an all-optical quantitative phase microscope (QPM) design that requires no computational post-reconstruction. A follow-up literature survey suggested that the learnt architecture is similar to the generalized phase contrast method developed two decades ago. Our extensive experiments on multiple datasets that include biological samples show that our learnt all-optical QPM designs consistently outperform existing methods. We experimentally verify the functionality of the optical 4-$f$ system based QPM design using a spatial light modulator. Furthermore, we also demonstrate that similar results can be achieved by an uninterpretable learning based method, namely diffractive deep neural networks (D2NN). The proposed differentiable microscopy framework supplements the creative process of designing new optical systems and would perhaps lead to unconventional but better optical designs.

Covid-19大流行导致了前所未有的全球公共卫生危机。鉴于其固有的性质，建议社会疏散措施作为遏制这种大流行传播的主要策略。因此，识别违反这些协议的情况，对削减疾病的传播并促进可持续生活方式具有影响。本文提出了一种基于电脑视觉的基于计算机视觉的系统，分析了CCTV镜头，为Covid-19传播提供了威胁水平评估。该系统努力捕获跨越多个帧的CCTV镜头的信息内容，以识别各个帧的各种违反社会偏移协议的实例，以及跨空间的识别，以及组行为的识别。该功能主要是通过利用基于时间图的基础结构来实现CCTV镜头的信息和对全能解释图的策略并量化给定场景的威胁级别的策略。在一系列场景中测试并验证各个组件，并针对人类专家意见进行了完整的系统。结果反映了威胁水平对人，其物理接近，相互作用，防护服和群体动力学的依赖。系统性能的准确性为76％，从而在城市进行了可部署的威胁监控系统，以允许社会中的正常和可持续性。