419,506 research outputs found
Player agency in interactive narrative: audience, actor & author
Get PDFThe question motivating this review paper is, how can
computer-based interactive narrative be used as a constructivist learn-
ing activity? The paper proposes that player agency can be used to
link interactive narrative to learner agency in constructivist theory,
and to classify approaches to interactive narrative. The traditional
question driving research in interactive narrative is, ‘how can an in-
teractive narrative deal with a high degree of player agency, while
maintaining a coherent and well-formed narrative?’ This question
derives from an Aristotelian approach to interactive narrative that,
as the question shows, is inherently antagonistic to player agency.
Within this approach, player agency must be restricted and manip-
ulated to maintain the narrative. Two alternative approaches based
on Brecht’s Epic Theatre and Boal’s Theatre of the Oppressed are
reviewed. If a Boalian approach to interactive narrative is taken the
conflict between narrative and player agency dissolves. The question
that emerges from this approach is quite different from the traditional
question above, and presents a more useful approach to applying in-
teractive narrative as a constructivist learning activity
Designing for usability of an adaptive time management assistant
No full textThis case study article describes the iterative design process of an AI-based mixed-initiative calendaring tool with embedded artificial intelligence. We establish the specific types of assistance in which the target user population expressed interest, and we highlight our findings regarding the scheduling practices and the reminding preferences of these users. These findings motivated the redesign and enhancement of our intelligent system. Lessons learned from the study-namely, that AI systems must be usable to gain widespread adoption and retention and that simple problems that perhaps do not necessitate complex AI-based solutions should not go unattended merely because of their inherent simplicity-conclude the article, along with a discussion of the importance of the iterative design process for any user adaptive system. Copyright © 2009, Association for the Advancement of Artificial Intelligence. All rights reserved.BERRY P, 2009, P 21 C INN APPL ART; BERRY R, 2006, P 5 INT JOINT C AUT, P1564; GLASS A, 2004, P 13 INT C INT US IN, P227; GOULD JD, 1985, COMMUN ACM, V28, P300, DOI 10.1145-3166.3170; Kozierok Robyn, 1993, P ACM SIGCHI INT WOR, P81, DOI 10.1145-169891.169908; MOM P. J., 2004, P 6 INT BIC WORKSH A, P169; Myers K, 2007, AI MAG, V28, P47; Palen L. Social, 1999, P SIGCHI C HUM FACT, P17, DOI 10.1145-302979.302982; Rich C., 2005, Proceedings. 25th IEEE International Conference on Distributed Computing Systems Workshops; SPAULDING A, 2009, AI MAGAZINE, V30; Tullio J., 2002, P 15 ANN S US INT SO, P11; WEBER JS, 2008, 26 C HUM FACT COMP S, P294910
Systematic Mapping Study on the Machine Learning Lifecycle
No full textThe development of artificial intelligence (AI) has made various industries eager to explore the benefits of AI. There is an increasing amount of research surrounding AI, most of which is centred on the development of new AI algorithms and techniques. However, the advent of AI is bringing an increasing set of practical problems related to AI model lifecycle management that need to be investigated. We address this gap by conducting a systematic mapping study on the lifecycle of AI model. Through quantitative research, we provide an overview of the field, identify research opportunities, and provide suggestions for future research. Our study yields 405 publications published from 2005 to 2020, mapped in 5 different main research topics, and 31 sub-topics. We observe that only a minority of publications focus on data management and model production problems, and that more studies should address the AI lifecycle from a holistic perspective.Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Software EngineeringData-Intensive System
Learning Occupational Task-Shares Dynamics for the Future of Work
No full text© 2020 Copyright held by the owner/author(s). The recent wave of AI and automation has been argued to differ from previous General Purpose Technologies (GPTs), in that it may lead to rapid change in occupations' underlying task requirements and persistent technological unemployment. In this paper, we apply a novel methodology of dynamic task shares to a large dataset of online job postings to explore how exactly occupational task demands have changed over the past decade of AI innovation, especially across high, mid and low wage occupations. Notably, big data and AI have risen significantly among high wage occupations since 2012 and 2016, respectively. We built an ARIMA model to predict future occupational task demands and showcase several relevant examples in Healthcare, Administration, and IT. Such task demands predictions across occupations will play a pivotal role in retraining the workforce of the future
Non-imaging Medical Data Synthesis for Trustworthy AI: A Comprehensive Survey
Get PDFPublisher Copyright: © 2024 Copyright held by the owner/author(s).Data quality is a key factor in the development of trustworthy AI in healthcare. A large volume of curated datasets with controlled confounding factors can improve the accuracy, robustness, and privacy of downstream AI algorithms. However, access to high-quality datasets is limited by the technical difficulties of data acquisition, and large-scale sharing of healthcare data is hindered by strict ethical restrictions. Data synthesis algorithms, which generate data with distributions similar to real clinical data, can serve as a potential solution to address the scarcity of good quality data during the development of trustworthy AI. However, state-of-the-art data synthesis algorithms, especially deep learning algorithms, focus more on imaging data while neglecting the synthesis of non-imaging healthcare data, including clinical measurements, medical signals and waveforms, and electronic healthcare records (EHRs). Therefore, in this article, we will review synthesis algorithms, particularly for non-imaging medical data, with the aim of providing trustworthy AI in this domain. This tutorial-style review article will provide comprehensive descriptions of non-imaging medical data synthesis, covering aspects such as algorithms, evaluations, limitations, and future research directions.Peer reviewe
AI management beyond the hype : exploring the co-constitution of AI and organizational context
No full textAI technologies hold great promise for addressing existing problems in organizational contexts, but the potential benefits must not obscure the potential perils associated with AI. In this article, we conceptually explore these promises and perils by examining AI use in organizational contexts. The exploration complements and extends extant literature on AI management by providing a typology describing four types of AI use, based on the idea of co-constitution of AI technologies and organizational context. Building on this typology, we propose three recommendations for informed use of AI in contemporary organizations. First, explicitly define the purpose of organizational AI use. Second, define the appropriate level of transparency and algorithmic management for organizational AI use. Third, be aware of AI's context-dependent nature
AI vs. AI: Can AI Detect AI-Generated Images?
No full textThe proliferation of Artificial Intelligence (AI) models such as Generative Adversarial Networks (GANs) has shown impressive success in image synthesis. Artificial GAN-based synthesized images have been widely spread over the Internet with the advancement in generating naturalistic and photo-realistic images. This might have the ability to improve content and media; however, it also constitutes a threat with regard to legitimacy, authenticity, and security. Moreover, implementing an automated system that is able to detect and recognize GAN-generated images is significant for image synthesis models as an evaluation tool, regardless of the input modality. To this end, we propose a framework for reliably detecting AI-generated images from real ones through Convolutional Neural Networks (CNNs). First, GAN-generated images were collected based on different tasks and different architectures to help with the generalization. Then, transfer learning was applied. Finally, several Class Activation Maps (CAM) were integrated to determine the discriminative regions that guided the classification model in its decision. Our approach achieved 100% on our dataset, i.e., Real or Synthetic Images (RSI), and a superior performance on other datasets and configurations in terms of its accuracy. Hence, it can be used as an evaluation tool in image generation. Our best detector was a pre-trained EfficientNetB4 fine-tuned on our dataset with a batch size of 64 and an initial learning rate of 0.001 for 20 epochs. Adam was used as an optimizer, and learning rate reduction along with data augmentation were incorporated
Barcoding of the cytochrome oxidase I (COI) indicates a recent introduction of Ciona savignyi into New Zealand and provides a rapid method for Ciona species discrimination
No full textMitochondrial cytochrome oxidase I (COI) gene sequencing (DNA barcoding) of Ciona specimens from New Zealand (NZ) led to the first record of the solitary ascidian Ciona savignyi in the Southern Hemisphere. We sought to quantify C. savignyi COI genetic diversity around the NZ archipelago and to compare this with diversity within C. savignyi's native range in the north-west Pacific. Ciona savignyi specimens were collected from two NZ sites and from three sites around Japan. COI sequences (595 bp) were amplified and measures of genetic diversity were calculated. Based on differences between their COI sequences we developed a PCR-based assay to distinguish C. savignyi from the morphologically similar C. intestinalis. A total of 12 C. savignyi COI haplotypes were recovered from the 76 samples. Of the four haplotypes observed in NZ, two were unique. From the 10 haplotypes observed in the Japan samples, eight were unique. The C. savignyi populations in Japan were found to contain higher haplotype diversity when compared with those in NZ. The NZ samples contained only a small subset of the haplotype variation of the Japan samples, however, NZ samples did harbor two haplotypes not observed in the Japan samples. A PCR-based assay developed from the COI sequences was able to reliably discriminate the two Ciona species. The low COI genetic diversity within the two NZ C. savignyi populations sampled is consistent with a founder effect associated loss of genetic diversity. The robust PCR-based assay for distinguishing C. savignyi and C. intestinalis may find application in ecological and taxonomic studies and can be applied to both archival materials and live animals
Is AI ground truth really true? The dangers of training and evaluating AI tools based on experts’ know-what
No full textOrganizational decision-makers need to evaluate AI tools in light of increasing claims that such tools outperform human experts. Yet, measuring the quality of knowledge work is challenging, raising the question of how to evaluate AI performance in such contexts. We investigate this question through a field study of a major U.S. hospital, observing how managers evaluated five different machine-learning (ML) based AI tools. Each tool reported high performance according to standard AI accuracy measures, which were based on ground truth labels provided by qualified experts. Trying these tools out in practice, however, revealed that none of them met expectations. Searching for explanations, managers began confronting the high uncertainty of experts’ know-what knowledge captured in ground truth labels used to train and validate ML models. In practice, experts address this uncertainty by drawing on rich know-how practices, which were not incorporated into these ML-based tools. Discovering the disconnect between AI’s know-what and experts’ know-how enabled managers to better understand the risks and benefits of each tool. This study shows dangers of treating ground truth labels used in ML models objectively when the underlying knowledge is uncertain. We outline implications of our study for developing, training, and evaluating AI for knowledge work
AI-Native Communications
No full textThe emergence of artificial intelligence (AI)-based methods evolving from 5G to 6G is accelerating. Therefore, to optimize the communication system in the 6G era, it is essential to adapt several AI-based optimization methods according to each environment. In this chapter, we introduce two general AI-based optimization methods, named AI-aided and AI-native. In addition, we describe the pros and cons of each method. Finally, by illustrating previous studies on AI in communication, we aim to speed up the development of both AI-based and AI-native optimization methods in line with the upcoming 6G era
- …
