Reliable and frequent population estimation is key for making policies around vaccination and planning infrastructure delivery. Since censuses lack the spatio-temporal resolution required for these tasks, census-independent approaches, using remote sensing and microcensus data, have become popular. We estimate intercensal population count in two pilot districts in Mozambique. To encourage sustainability, we assess the feasibility of using publicly available datasets to estimate population. We also explore transfer learning with existing annotated datasets for predicting building footprints, and training with additional `dot' annotations from regions of interest to enhance these estimations. We observe that population predictions improve when using footprint area estimated with this approach versus only publicly available features.

INTRODUCTION With increasing awareness of the societal risks of algorithmic bias and encroaching automation, issues of fairness, accountability, and transparency in data-driven AI systems have received growing academic attention in multiple high-stakes contexts, including healthcare, loan-granting, and hiring (e.g., Barocas & Selbst, 2016; Holstein, Wortman Vaughan, Daumé III, Dudik, & Wallach, 2019; Veale, Van Kleek, & Binns, 2018). Given these noble intentions, why might AIEd systems have inequitable impacts? In this chapter, we ask whether AIEd systems will ultimately serve to A mplify I nequities in Ed ucation, or alternatively, whether they will help to A lleviate existing inequities. We discuss four lenses that can be used to examine how and why AIEd systems risk amplifying existing inequities: (1) factors inherent to the overall socio-technical system design; (2) the use of datasets that reflect historical inequities; (3) factors inherent to the underlying algorithms used to drive machine learning and automated decision-making, and (4) factors that emerge through a complex interplay between automated and human decision-making. Building from these lenses, we then outline possible paths towards more equitable futures for AIEd, while highlighting debates surrounding each proposal. In doing so, we hope to provoke new conversations around the design of equitable AIEd, and to push ongoing conversations in the field forward. PATHWAYS TOWARD INEQUITY IN AIED We begin by presenting four lenses to understand how AIEd systems might amplify existing inequities or even create new ones (cf. While each lens provides a different way of examining pathways towards inequity in AIEd, all are pointed at the same underlying socio-technical system. Figure 1 provides a coarse-grained overview of the broader social-technical systems in which AIEd systems are embedded, and some of the components we will refer to in the four lenses. The accumulated, collective decisions of designers, researchers, policy-makers, and other stakeholders shape these systems' designs. In addition to using or being affected by AIEd systems, on-the-ground stakeholders such as students, teachers, or school administrators may also play a role in shaping their designs; whether directly, through participatory design processes, or indirectly through the passive generation of training data while interacting with an AIEd interface. In turn, decisions regarding what data is used to shape an AIEd system's design (e.g., when used as training data for use with machine learning methods) can shape an AIEd system's algorithmic behavior (e.g., instructional policies learned from data).

To understand these manipulations of the quantum world, we need to understand the... Harvey pulled Einstein's brain out of the skull. Then put it in a formalin jar.... To understand these manipulations of the quantum world, we need to understand the... What is the most coveted and prestigious award in the world? Scientists recently developed this new process in the lab. Let's take a look at what Bill Gates was saying about the epidemic that we have... The whole world is now pointing to the Chinese wildlife market as a possible source... Today we are going to discuss Marie Tharp and Marie Tharp's historical map.

If there's one thing the United States military gets right, it's lethality. Yet even once the US military has you in its sights, it may not know who you actually are -- such are, these so-called "signature strikes" -- even as that wrathful finger of God is called down from upon on high. As Kate Crawford, Microsoft Research principal and co-founder of the AI Now Institute at NYU, lays out in this fascinating excerpt from her new book, Atlas of AI, the military-industrial complex is alive and well and now leveraging metadata surveillance scores derived by IBM to decide which home/commute/gender reveal party to drone strike next. And if you think that same insidious technology isn't already trickling down to infest the domestic economy, I have a credit score to sell you. Excerpted from Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence by Kate Crawford, published by Yale University Press.

The study of the social impact of automated decision making has focused largely on issues of fairness at the point of decision, evaluating the fairness (with respect to a population) of a sequence or pipeline of decisions, or examining the dynamics of a game between the decision-maker and the decision subject. What is missing from this study is an examination of precarity: a term coined by Judith Butler to describe an unstable state of existence in which negative decisions can have ripple effects on one's well-being. Such ripple effects are not captured by changes in income or wealth alone or by one decision alone. To study precarity, we must reorient our frame of reference away from the decision-maker and towards the decision subject; away from aggregates of decisions over a population and towards aggregates of decisions (for an individual) over time. An individual who lives with higher precarity is more affected and less able to recover by the same negative decision than another with low precarity. Thus including only the direct impact of a single decision or a few decisions is insufficient to judge if that system was fair. However, precarity is not an attribute of an individual; it is a result of being subject to greater risks and fewer supports, in addition to starting off at a less secure position. Precarity is impacted by racism, sexism, ableism, heterosexism, and other systems of oppression, and an individual's intersectional identity may put one at greater risk in society, subject to a lower income for the same job, less able to build wealth even at the same income level, and less able to recover from harm.

During the first weeks of February, we asked recipients of our Data and AI Newsletters to participate in a survey on AI adoption in the enterprise. We were interested in answering two questions. First, we wanted to understand how the use of AI grew in the past year. We were also interested in the practice of AI: how developers work, what techniques and tools they use, what their concerns are, and what development practices are in place. The most striking result is the sheer number of respondents. In our 2020 survey, which reached the same audience, we had 1,239 responses. This year, we had a total of 5,154. After eliminating 1,580 respondents who didn't complete the survey, we're left with 3,574 responses--almost three times as many as last year.

In this paper, we present a new objective prediction model for synthetic speech naturalness. It can be used to evaluate Text-To-Speech or Voice Conversion systems and works language independently. The model is trained end-to-end and based on a CNN-LSTM network that previously showed to give good results for speech quality estimation. We trained and tested the model on 16 different datasets, such as from the Blizzard Challenge and the Voice Conversion Challenge. Further, we show that the reliability of deep learning-based naturalness prediction can be improved by transfer learning from speech quality prediction models that are trained on objective POLQA scores. The proposed model is made publicly available and can, for example, be used to evaluate different TTS system configurations.

Currently, the most widespread neural network architecture for training language models is the so called BERT which led to improvements in various NLP tasks. In general, the larger the number of parameters in a BERT model, the better the results obtained in these NLP tasks. Unfortunately, the memory consumption and the training duration drastically increases with the size of these models, though. In this article, we investigate various training techniques of smaller BERT models and evaluate them on five public German NER tasks of which two are introduced by this article. We combine different methods from other BERT variants like ALBERT, RoBERTa, and relative positional encoding. In addition, we propose two new fine-tuning techniques leading to better performance: CSE-tagging and a modified form of LCRF. Furthermore, we introduce a new technique called WWA which reduces BERT memory usage and leads to a small increase in performance.

Deep probabilistic generative models have achieved incredible success in many fields of application. Among such models, variational autoencoders (VAEs) have proved their ability in modeling a generative process by learning a latent representation of the input. In this paper, we propose a novel VAE defined in the quaternion domain, which exploits the properties of quaternion algebra to improve performance while significantly reducing the number of parameters required by the network. The success of the proposed quaternion VAE with respect to traditional VAEs relies on the ability to leverage the internal relations between quaternion-valued input features and on the properties of second-order statistics which allow to define the latent variables in the augmented quaternion domain. In order to show the advantages due to such properties, we define a plain convolutional VAE in the quaternion domain and we evaluate its performance with respect to its real-valued counterpart on the CelebA face dataset.

Vulnerability detection has always been the most important task in the field of software security. With the development of technology, in the face of massive source code, automated analysis and detection of vulnerabilities has become a current research hotspot. For special text files such as source code, using some of the hottest NLP technologies to build models and realize the automatic analysis and detection of source code has become one of the most anticipated studies in the field of vulnerability detection. This article does a brief survey of some recent new documents and technologies, such as CodeBERT, and summarizes the previous technologies.