10530 research outputs found
Sort by
Hyperbolic Adversarial Learning for Personalized Item Recommendation
Get PDFPersonalized recommendation systems are indispensable intelligent components for social media and e-commerce. Traditional personalized item recommendation models are vulnerable to adversarial perturbations, resulting in poor robustness. Although adversarial learning-based recommendation models are able to improve the robustness, they inherently model the interaction relationships between users and items in Euclidean space, where it is difficult for them to capture the hierarchical relationships among entities. To address the above issues, we propose a hyperbolic adversarial learning based personalized item recommendation model, called HALRec. Specifically, HALRec models the interactions in hyperbolic space and utilizes hyperbolic distances to measure the similarities among entities. Moreover, instead of in Euclidean space, HALRec exploits the adversarial learning technique in hyperbolic space, i.e., HAL-Rec maximizes the hyperbolic adversarial perturbations loss while minimizing the hyperbolic based Bayesian personalized ranking loss. Hence, HALRec inherits the advantages of hyperbolic representation learning in capturing hierarchical relationships and adversarial learning in enhancing the robustness of the recommendation model. In addition, we utilize tangent space optimization to simplify the learning of model parameters. Experimental results on real-world datasets show that our proposed hyperbolic adversarial learning-based personalized item recommendation method outperforms the state-of-the-art personalized recommendation algorithms
Hum/Ine: Interspecies Empathy, Composition, and Horses
Get PDFThis project explores the relationships between horses and humans, primarily focussing on the empathetic aspects of sound and communication. The research investigates how working with sound can deepen our understanding of and connection with horses and, how a creative compositional practice could provide an opportunity to enhance interspecies bonds and interactions.The portfolio uses a combination of methodologies to explore these ideas of interspecies empathy. Key elements include: sound and listening technologies (including binaural microphones) that allow for detailed sonic capture; the idea of creative interspecies collaboration; the creative implications of zoomusicology; and the importance of listening with- as well as to- animals. The research outputs take the form of electroacoustic compositions, text-based scores, and an installation. In encouraging horses to participate in the process, the work aims to investigate or facilitate interspecies communication, co-creation/collaboration, and extra-linguistic communication.The pieces in the portfolio do not seek to replicate the listening experience of the horse but are proposals for how we might creatively approximate such experiences to explore ideas of interspecies empathy. The research is inherently multidisciplinary and provides novel methods of exploring equine experiences and empathetic connections through creative and playful means. The work offers new frameworks for collaborative and co-creative art practices involving non-human participants. In the portfolio I present innovative approaches which attempt to blend artistic practice with scientific curiosity
The phonological store of working memory:A critique and an alternative, perceptual-motor, approach to verbal short-term memory
Get PDFA key quality of a good theory is its fruitfulness, one measure of which might be the degree to which it compels researchers to test it, refine it, or offer alternative explanations of the same empirical data. Perhaps the most fruitful element of Baddeley and Hitch’s (1974) Working Memory framework has been the concept of a short-term phonological store, a discrete cognitive module dedicated to the passive storage of verbal material that is architecturally fractionated from perceptual, language, and articulatory systems. This review discusses how the phonological store construct has served as the main theoretical springboard for an alternative perceptual-motor approach in which serial recall performance reflects the opportunistic co-opting of the articulatory planning system and, when auditory material is involved, the products of obligatory auditory perceptual organisation. It is argued that this approach, which rejects the need to posit a distinct short-term store, provides a better account of the two putative empirical hallmarks of the phonological store—the phonological similarity effect and the irrelevant speech effect—and that it shows promise too in being able to account for nonword repetition and word-form learning, the supposed evolved function of the phonological store. The neuropsychological literature cited as strong additional support for the phonological store concept is also scrutinised through the lens of the perceptual-motor approach for the first time and a tentative articulatory-planning deficit hypothesis for the ‘short-term memory’ patient profile is advanced. Finally, the relation of the perceptual-motor approach to other ‘emergent-property’ accounts of short-term memory is briefly considered
Strategic narrative and public diplomacy:What Artificial Intelligence Means for the Endless Problem of Plural Meanings of Plural Things
Get PDFThis chapter advances a narrative approach to the study of public diplomacy. Webring together two phenomena: information disorder in communication, and order in world politics, to examine the challenges of narrating public diplomacy. We examine how actors can use tools of information disorder to further their strategic aims to shape international order. We do this in several ways. First, we set out these two (dis)order phenomena and their relationship. Second, we set out the dilemma of establishing and verifying truth claims in this information disorder. Third, we demonstrate why analysis of actors’ strategic narratives used in this context can explain how they are using information disorder to further their claims. Fourth and finally, we explore how generative artificial intelligence (AI) offers new tools for communication in foreign policy. It is important to examine both how actors use these tools, and how they try to control and direct the development of these tools. We argue that these tools add another dimension to a contested multipolarinternational order, one that extends a basic problem that generates politics: different people in different places prioritise different things and give things different meanings. Generative AI will not change this or solve this. This means an increasing complexity of communication since we wrote of strategic narrative in 2010. However, the distinct practices of actors using narratives to shape behaviour, and narratives being fundamental to how citizens view the world, remains unchanged
Redrawing Democracy:Quantifying House District Continuity and Change, 1789-2024
Get PDFHow have U.S. House districts changed since the Founding? When do mapmakers reshape districts - and when do they choose to preserve them? This article introduces the concept of 'spatial protectionism' - the strategic minimization of district change - and develops four original measures to analyze redistricting changes from 1789 onwards. I demonstrate how the strategic imperatives, racial context, and technical capabilities of mapmakers shape the magnitude and form of district change across five redistricting eras: Spatial Representation, Shifting Apportionment, Relative Stasis, Racial Redistricting, and Precision Engineering. My analysis reveals persistent low change disrupted by shifts to at-large districts; unprecedented boundary changes in the last decade; and clear associations between racial demographics, party control, and patterns of continuity and change. These findings establish the first empirical standard for judging the magnitude of district change and underscore the dual significance of stasis and dynamism in the politics of redistricting
Data sources:
Access Repository Dashboard
Do you manage Royal Holloway - Pure? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!