Following the onset of the COVID-19 pandemic and subsequent lockdowns, software engineers' daily life was disrupted and they were abruptly forced into working remotely from home. Across one exploratory and one confirmatory study (N = 482), we tested whether a typical working day is different to pre-pandemic times and whether specific tasks are associated with task-specific satisfaction and productivity. To explore the subject domain, we first run a two-wave longitudinal study, where we found that the time software engineers spent doing specific tasks (e.g., coding, bugfixing, helping others) from home was similar to pre-pandemic times. Also, the amount of time developers spent on each task was unrelated to their general well-being, perceived productivity, and other variables such as basic needs. In our confirmatory study, we found that task satisfaction and productivity are predicted by task-specific variables (e.g., how much autonomy software engineers had during coding) but not by task-independent variables such as general resilience or a good work-life balance. Additionally, we found that satisfaction and autonomy were significantly higher when software engineers were helping others and lower when they were bugfixing. Also, contrary to anecdotal evidence, software engineers' satisfaction and productivity during meetings is not lower compared to other tasks. Finally, we discuss implications for software engineers, management, and researchers.
相關內容
This paper describes the process of developing data visualisations to enhance a commercial software platform for combating insider threat, whose existing UI, while perfectly functional, was limited in its ability to allow analysts to easily spot the patterns and outliers that visualisation naturally reveals. We describe the design and development process, proceeding from initial tasks/requirements gathering, understanding the platform's data formats, the rationale behind the visualisation's design, and then refining the prototype through gathering feedback from representative domain experts who are also current users of the software. Through a number of example scenarios, we show that the visualisation can support the identified tasks and aid analysts in discovering and understanding potentially risky insider activity within a large user base.
Much software, whether beneficent or malevolent, is distributed only as binaries, sans source code. Absent source code, understanding binaries' behavior can be quite challenging, especially when compiled under higher levels of compiler optimization. These optimizations can transform comprehensible, "natural" source constructions into something entirely unrecognizable. Reverse engineering binaries, especially those suspected of being malevolent or guilty of intellectual property theft, are important and time-consuming tasks. There is a great deal of interest in tools to "decompile" binaries back into more natural source code to aid reverse engineering. Decompilation involves several desirable steps, including recreating source-language constructions, variable names, and perhaps even comments. One central step in creating binaries is optimizing function calls, using steps such as inlining. Recovering these (possibly inlined) function calls from optimized binaries is an essential task that most state-of-the-art decompiler tools try to do but do not perform very well. In this paper, we evaluate a supervised learning approach to the problem of recovering optimized function calls. We leverage open-source software and develop an automated labeling scheme to generate a reasonably large dataset of binaries labeled with actual function usages. We augment this large but limited labeled dataset with a pre-training step, which learns the decompiled code statistics from a much larger unlabeled dataset. Thus augmented, our learned labeling model can be combined with an existing decompilation tool, Ghidra, to achieve substantially improved performance in function call recovery, especially at higher levels of optimization.
Machine learning (ML) provides us with numerous opportunities, allowing ML systems to adapt to new situations and contexts. At the same time, this adaptability raises uncertainties concerning the run-time product quality or dependability, such as reliability and security, of these systems. Systems can be tested and monitored, but this does not provide protection against faults and failures in adapted ML systems themselves. We studied software designs that aim at introducing fault tolerance in ML systems so that possible problems in ML components of the systems can be avoided. The research was conducted as a case study, and its data was collected through five semi-structured interviews with experienced software architects. We present a conceptualisation of the misbehaviour of ML systems, the perceived role of fault tolerance, and the designs used. Common patterns to incorporating ML components in design in a fault tolerant fashion have started to emerge. ML models are, for example, guarded by monitoring the inputs and their distribution, and enforcing business rules on acceptable outputs. Multiple, specialised ML models are used to adapt to the variations and changes in the surrounding world, and simpler fall-over techniques like default outputs are put in place to have systems up and running in the face of problems. However, the general role of these patterns is not widely acknowledged. This is mainly due to the relative immaturity of using ML as part of a complete software system: the field still lacks established frameworks and practices beyond training to implement, operate, and maintain the software that utilises ML. ML software engineering needs further analysis and development on all fronts.
We conduct a large-scale social media-based study of oral health during the COVID-19 pandemic based on tweets from 9,104 Twitter users across 26 states (with sufficient samples) in the United States for the period between November 12, 2020 and June 14, 2021. To better understand how discussions on different topics/oral diseases vary across the users, we acquire or infer demographic information of users and other characteristics based on retrieved information from user profiles. Women and younger adults (19-29) are more likely to talk about oral health problems. We use the LDA topic model to extract the major topics/oral diseases in tweets. Overall, 26.70% of the Twitter users talk about wisdom tooth pain/jaw hurt, 23.86% tweet about dental service/cavity, 18.97% discuss chipped tooth/tooth break, 16.23% talk about dental pain, and the rest are about tooth decay/gum bleeding. By conducting logistic regression, we find that discussions vary across user characteristics. More importantly, we find social disparities in oral health during the pandemic. Specifically, we find that health insurance coverage rate is the most significant predictor in logistic regression for topic prediction. People from counties with higher insurance coverage tend to tweet less about all topics of oral diseases. People from counties at a higher risk of COVID-19 talk more about tooth decay/gum bleeding and chipped tooth/tooth break. Older adults (50+), who are vulnerable to COVID-19, are more likely to discuss dental pain. To our best knowledge, this is the first large-scale social media-based study to analyze and understand oral health in America amid the COVID-19 pandemic. We hope the findings of our study through the lens of social media can provide insights for oral health practitioners and policy makers.
We investigate the impact of the COVID-19 pandemic on the betting markets of professional and college sports. We find that during the pandemic, the moneyline betting markets of the National Basketball Association (NBA) became very inefficient. During this period, if one bet uniformly on underdog teams in the NBA, one could achieve a 16.7% profit margin. It is hypothesized that this inefficiency is due to the absence of live audiences during the NBA games. Such inefficiencies are not seen for any other sport. Much of the inefficiency comes from a small fraction of games with moneyline odds in the 233 to 400 range. We find that with clever strategies, one is able to achieve a 26-fold gain on an initial investment by betting on NBA underdogs during this time period.
Fact-checking has become increasingly important due to the speed with which both information and misinformation can spread in the modern media ecosystem. Therefore, researchers have been exploring how fact-checking can be automated, using techniques based on natural language processing, machine learning, knowledge representation, and databases to automatically predict the veracity of claims. In this paper, we survey automated fact-checking stemming from natural language processing, and discuss its connections to related tasks and disciplines. In this process, we present an overview of existing datasets and models, aiming to unify the various definitions given and identify common concepts. Finally, we highlight challenges for future research.
Clustering is one of the most fundamental and wide-spread techniques in exploratory data analysis. Yet, the basic approach to clustering has not really changed: a practitioner hand-picks a task-specific clustering loss to optimize and fit the given data to reveal the underlying cluster structure. Some types of losses---such as k-means, or its non-linear version: kernelized k-means (centroid based), and DBSCAN (density based)---are popular choices due to their good empirical performance on a range of applications. Although every so often the clustering output using these standard losses fails to reveal the underlying structure, and the practitioner has to custom-design their own variation. In this work we take an intrinsically different approach to clustering: rather than fitting a dataset to a specific clustering loss, we train a recurrent model that learns how to cluster. The model uses as training pairs examples of datasets (as input) and its corresponding cluster identities (as output). By providing multiple types of training datasets as inputs, our model has the ability to generalize well on unseen datasets (new clustering tasks). Our experiments reveal that by training on simple synthetically generated datasets or on existing real datasets, we can achieve better clustering performance on unseen real-world datasets when compared with standard benchmark clustering techniques. Our meta clustering model works well even for small datasets where the usual deep learning models tend to perform worse.
There has been considerable growth and interest in industrial applications of machine learning (ML) in recent years. ML engineers, as a consequence, are in high demand across the industry, yet improving the efficiency of ML engineers remains a fundamental challenge. Automated machine learning (AutoML) has emerged as a way to save time and effort on repetitive tasks in ML pipelines, such as data pre-processing, feature engineering, model selection, hyperparameter optimization, and prediction result analysis. In this paper, we investigate the current state of AutoML tools aiming to automate these tasks. We conduct various evaluations of the tools on many datasets, in different data segments, to examine their performance, and compare their advantages and disadvantages on different test cases.
It is becoming increasingly easy to automatically replace a face of one person in a video with the face of another person by using a pre-trained generative adversarial network (GAN). Recent public scandals, e.g., the faces of celebrities being swapped onto pornographic videos, call for automated ways to detect these Deepfake videos. To help developing such methods, in this paper, we present the first publicly available set of Deepfake videos generated from videos of VidTIMIT database. We used open source software based on GANs to create the Deepfakes, and we emphasize that training and blending parameters can significantly impact the quality of the resulted videos. To demonstrate this impact, we generated videos with low and high visual quality (320 videos each) using differently tuned parameter sets. We showed that the state of the art face recognition systems based on VGG and Facenet neural networks are vulnerable to Deepfake videos, with 85.62% and 95.00% false acceptance rates respectively, which means methods for detecting Deepfake videos are necessary. By considering several baseline approaches, we found that audio-visual approach based on lip-sync inconsistency detection was not able to distinguish Deepfake videos. The best performing method, which is based on visual quality metrics and is often used in presentation attack detection domain, resulted in 8.97% equal error rate on high quality Deepfakes. Our experiments demonstrate that GAN-generated Deepfake videos are challenging for both face recognition systems and existing detection methods, and the further development of face swapping technology will make it even more so.
This paper surveys the current state of the art in Natural Language Generation (NLG), defined as the task of generating text or speech from non-linguistic input. A survey of NLG is timely in view of the changes that the field has undergone over the past decade or so, especially in relation to new (usually data-driven) methods, as well as new applications of NLG technology. This survey therefore aims to (a) give an up-to-date synthesis of research on the core tasks in NLG and the architectures adopted in which such tasks are organised; (b) highlight a number of relatively recent research topics that have arisen partly as a result of growing synergies between NLG and other areas of artificial intelligence; (c) draw attention to the challenges in NLG evaluation, relating them to similar challenges faced in other areas of Natural Language Processing, with an emphasis on different evaluation methods and the relationships between them.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191