More Like this

1. The effect of model structure and data availability on Usutu virus dynamics at three biological scales

   https://doi.org/10.1098/rsos.231146  

   Heitzman-Breen, Nora; Liyanage, Yuganthi R; Duggal, Nisha; Tuncer, Necibe; Ciupe, Stanca M (February 2024, Royal Society Open Science)

   Understanding the epidemiology of emerging pathogens, such as Usutu virus (USUV) infections, requires systems investigation at each scale involved in the host–virus transmission cycle, from individual bird infections, to bird-to-vector transmissions, and to USUV incidence in bird and vector populations. For new pathogens field data are sparse, and predictions can be aided by the use of laboratory-type inoculation and transmission experiments combined with dynamical mathematical modelling. In this study, we investigated the dynamics of two strains of USUV by constructing mathematical models for the within-host scale, bird-to-vector transmission scale and vector-borne epidemiological scale. We used individual within-host infectious virus data and per cent mosquito infection data to predict USUV incidence in birds and mosquitoes. We addressed the dependence of predictions on model structure, data uncertainty and experimental design. We found that uncertainty in predictions at one scale change predicted results at another scale. We proposedin silicoexperiments that showed that sampling every 12 hours ensures practical identifiability of the within-host scale model. At the same time, we showed that practical identifiability of the transmission scale functions can only be improved under unrealistically high sampling regimes. Instead, we proposed optimal experimental designs and suggested the types of experiments that can ensure identifiability at the transmission scale and, hence, induce robustness in predictions at the epidemiological scale. 

   more » « less
2. A dataset for the study of identity at scale: Annual Prevalence of American Twitter Users with specified Token in their Profile Bio 2015–2020

   https://doi.org/10.1371/journal.pone.0260185  

   Jones, Jason Jeffrey (November 2021, PLOS ONE)

   Cotfas, Liviu-Adrian (Ed.)

   Personally expressed identity is who or what an individual themselves says they are, and it should be studied at scale. At scale means with data on millions of individuals, which is newly available and comes timestamped and geocoded. This work introduces a dataset for the study of identity at scale and describes the method for collecting and aggregating such data. Further, tools and theory for working with the data are presented. A demonstration analysis provides evidence that personal, individual development and changing cultural norms can be observed with these data and methods. 

   more » « less
3. Auto-Differentiation of Relational Computations for Very Large Scale Machine Learning

   Tang, Yuxin; Ding, Zhimin; Jankov, Dimitrije; Yuan, Binhang; Bourgeois, Daniel; Jermaine, Chris (July 2023, Proceedings of Machine Learning Research (ICML))

   The relational data model was designed to facilitate large-scale data management and analytics. We consider the problem of how to differentiate computations expressed relationally. We show experimentally that a relational engine running an auto-differentiated relational algorithm can easily scale to very large datasets, and is competitive with state-of-the-art, special-purpose systems for large-scale distributed machine learning. 

   more » « less
4. Auto-Differentiation of Relational Computations for Very Large Scale Machine Learning

   Yuxin Tang, Zhimin Ding (December 2023, Proceedings of Machine Learning Research)

   The relational data model was designed to facilitate large-scale data management and analytics. We consider the problem of how to differentiate computations expressed relationally. We show experimentally that a relational engine running an auto-differentiated relational algorithm can easily scale to very large datasets, and is competitive with state-of-the-art, special-purpose systems for large-scale distributed machine learning. 

   more » « less
5. Multi-Scale Graph Learning for Anti-Sparse Downscaling

   https://doi.org/10.1609/aaai.v39i27.35014  

   Fan, Yingda; Yu, Runlong; Barclay, Janet R; Appling, Alison P; Sun, Yiming; Xie, Yiqun; Jia, Xiaowei (April 2025, Proceedings of the AAAI Conference on Artificial Intelligence)

   Water temperature can vary substantially even across short distances within the same sub-watershed. Accurate prediction of stream water temperature at fine spatial resolutions (i.e., fine scales, ≤ 1 km) enables precise interventions to maintain water quality and protect aquatic habitats. Although spatiotemporal models have made substantial progress in spatially coarse time series modeling, challenges persist in predicting at fine spatial scales due to the lack of data at that scale. To address the problem of insufficient fine-scale data, we propose a Multi-Scale Graph Learning (MSGL) method. This method employs a multi-task learning framework where coarse-scale graph learning, bolstered by larger datasets, simultaneously enhances fine-scale graph learning. Although existing multi-scale or multi-resolution methods integrate data from different spatial scales, they often overlook the spatial correspondences across graph structures at various scales. To address this, our MSGL introduces an additional learning task, cross-scale interpolation learning, which leverages the hydrological connectedness of stream locations across coarse- and fine-scale graphs to establish cross-scale connections, thereby enhancing overall model performance. Furthermore, we have broken free from the mindset that multi-scale learning is limited to synchronous training by proposing an Asynchronous Multi-Scale Graph Learning method (ASYNC-MSGL). Extensive experiments demonstrate the state-of-the-art performance of our method for anti-sparse downscaling of daily stream temperatures in the Delaware River Basin, USA, highlighting its potential utility for water resources monitoring and management. 

   more » « less