More Like this

1. Leibniz International Proceedings in Informatics (LIPIcs):13th Innovations in Theoretical Computer Science Conference (ITCS 2022)

   https://doi.org/10.4230/LIPIcs.ITCS.2022.60  

   Ebrahimnejad, Farzam; Lee, James R. (January 2022, Schloss Dagstuhl – Leibniz-Zentrum für Informatik)

   Braverman, Mark (Ed.)

   We present an O((log n)²)-competitive algorithm for metrical task systems (MTS) on any n-point metric space that is also 1-competitive for service costs. This matches the competitive ratio achieved by Bubeck, Cohen, Lee, and Lee (2019) and the refined competitive ratios obtained by Coester and Lee (2019). Those algorithms work by first randomly embedding the metric space into an ultrametric and then solving MTS there. In contrast, our algorithm is cast as regularized gradient descent where the regularizer is a multiscale metric entropy defined directly on the metric space. This answers an open question of Bubeck (Highlights of Algorithms, 2019). 

   more » « less
2. Tephra Fusion 2022 workshop focused on best practices in tephra data recommends innovative computer solutions to build databases

   Wallace, K.L. (January 2022, EarthCube Annual Meeting 2022)

   A series of international workshops held in 2014, 2017, 2019, and 2022 focused on improving tephra studies from field collection through publication and encouraging FAIR (findable, accessible, interoperable, reusable) data practices for tephra data and metadata. Two consensus needs for tephra studies emerged from the 2014 and 2017 workshops: (a) standardization of tephra field data collection, geochemical analysis, correlation, and data reporting, and (b) development of next generation computer tools and databases to facilitate information access across multidisciplinary communities. To achieve (a), we developed a series of recommendations for best practices in tephra studies, from sample collection through analysis and data reporting (https://zenodo.org/record/3866266). A 4-part virtual workshop series (https://tephrochronology.org/cot/Tephra2022/) was held in February and March, 2022, to update the tephra community on these developments, to get community feedback, to learn of unmet needs, and to plan a future roadmap for open and FAIR tephra data. More than 230 people from 25 nations registered for the workshop series. The community strongly emphasized the need for better computer systems, including physical infrastructure (repositories and servers), digital infrastructure (software and tools) and human infrastructure (people, training, and professional assistance), to store, manage and serve global tephra datasets. Some desired attributes of improved computer systems include: 1) user friendliness 2) ability to easily ingest multiparameter tephra data (using best practice recommended data fields); 3) interoperability with existing data repositories; 4) development of tool add-ons (plotting and statistics); 5) improved searchability 6) development of a tephra portal with access to distributed data systems, and 7) commitments to long-term support from funding agencies, publishers and the cyberinfrastructure community. 

   more » « less
3. WACV 2022

   Matt Gorbett, Nathaniel Blanchard (January 2022, Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Workshops, 2022, pp. 34-43)

   Neural networks are vulnerable to a wide range of erroneous inputs such as corrupted, out-of-distribution, misclassified, and adversarial examples. Previously, separate solutions have been proposed for each of these faulty data types, however, in this work we show that a collective set of inputs with variegated data quality issues can be jointly identified with a single model. Specifically, we train a linear SVM classifier to detect four types of erroneous data using the hidden and softmax feature vectors of pre-trained neural networks. Our results indicate that these faulty data types generally exhibit linearly separable activation properties from correctly processed examples. We are able to identify erroneous inputs with an AUROC of 0.973 on CIFAR10, 0.957 on Tiny ImageNet, and 0.941 on ImageNet. We experimentally validate our findings across a diverse range of datasets, domains, and pre-trained models. 

   more » « less
4. InterPro in 2022

   https://doi.org/10.1093/nar/gkac993  

   Paysan-Lafosse, Typhaine; Blum, Matthias; Chuguransky, Sara; Grego, Tiago; Pinto, Beatriz Lázaro; Salazar, Gustavo A; Bileschi, Maxwell L; Bork, Peer; Bridge, Alan; Colwell, Lucy; et al (November 2022, Nucleic Acids Research)

   Abstract The InterPro database (https://www.ebi.ac.uk/interpro/) provides an integrative classification of protein sequences into families, and identifies functionally important domains and conserved sites. Here, we report recent developments with InterPro (version 90.0) and its associated software, including updates to data content and to the website. These developments extend and enrich the information provided by InterPro, and provide a more user friendly access to the data. Additionally, we have worked on adding Pfam website features to the InterPro website, as the Pfam website will be retired in late 2022. We also show that InterPro's sequence coverage has kept pace with the growth of UniProtKB. Moreover, we report the development of a card game as a method of engaging the non-scientific community. Finally, we discuss the benefits and challenges brought by the use of artificial intelligence for protein structure prediction. 

   more » « less
5. 2022 Annual Meeting

   https://doi.org/10.1785/0220220087  

   DUGDA, M.; KASSA, A. B.; POUCHARD, L.; DIRES, E.; MCDANIEL, L. (April 2022, Seismological Research Letters)

   Caplan-Auerbach, Jackie; Schmidt, D. (Ed.)

   Title: Efficient Access and Manipulation of Big Seismic Data from Disparate Sources Seismological data recordings have been growing exponentially in the last three decades. For instance, the data archive at the IRIS Data Management Center (DMC) grew from less than 10 Tebibytes in 1992 to greater than 750 Tebibytes today (in 2022). In addition to such big data archives, retrieving and merging data from various disparate seismic sources also creates big data which will enable obtaining higher-resolution seismic images and understanding phenomena such as earthquake cycles. Moreover, recent progress in geosciences with the application of AI/ML using big data has shown the potential of discovering patterns that were not previously recognized. However, aggregating large seismic datasets introduces its own challenges. Some of these challenges arise from the fact that many data centers have their own way of distributing data, and the format of the data and metadata are different in many cases. The objective of this investigation is the development of data access and manipulation tools for retrieval, merging, processing, and the management of big seismic data from disparate seismic data sources. We develop a free, open-source, direct data accessing, gathering, and processing software toolbox for disparate sources using Python. Aggregating data from different data centers will enable us to investigate the seismic structure beneath a region of interest at a higher resolution by merging the seismic databases. Such a merged dataset can be applied on studies around the boundaries between countries if those countries have different networks. One boundary region of geologic interest to exemplify the benefits of aggregated seismic datasets from different networks in two countries is the southern side of the Rio Grande Rift including the bordering areas between the US and Mexico. Previous more detailed seismic studies on Rio Grande were conducted mostly on the US side of the Rift Valley. 

   more » « less