Zobrazeno 1 - 10
of 4 593
pro vyhledávání: '"Bergsma, A"'
Per-example gradient norms are a vital ingredient for estimating gradient noise scale (GNS) with minimal variance. Observing the tensor contractions required to compute them, we propose a method with minimal FLOPs in 3D or greater tensor regimes by s
Externí odkaz:
http://arxiv.org/abs/2411.00999
Several challenges make it difficult for sparse neural networks to compete with dense models. First, setting a large fraction of weights to zero impairs forward and gradient signal propagation. Second, sparse studies often need to test multiple spars
Externí odkaz:
http://arxiv.org/abs/2405.15743
Developments and applications of the OPTIMADE API for materials discovery, design, and data exchange
Autor:
Evans, Matthew L., Bergsma, Johan, Merkys, Andrius, Andersen, Casper W., Andersson, Oskar B., Beltrán, Daniel, Blokhin, Evgeny, Boland, Tara M., Balderas, Rubén Castañeda, Choudhary, Kamal, Díaz, Alberto Díaz, García, Rodrigo Domínguez, Eckert, Hagen, Eimre, Kristjan, Montero, María Elena Fuentes, Krajewski, Adam M., Mortensen, Jens Jørgen, Duarte, José Manuel Nápoles, Pietryga, Jacob, Qi, Ji, Carrillo, Felipe de Jesús Trejo, Vaitkus, Antanas, Yu, Jusong, Zettel, Adam, de Castro, Pedro Baptista, Carlsson, Johan, Cerqueira, Tiago F. T., Divilov, Simon, Hajiyani, Hamidreza, Hanke, Felix, Jose, Kevin, Oses, Corey, Riebesell, Janosh, Schmidt, Jonathan, Winston, Donald, Xie, Christen, Yang, Xiaoyu, Bonella, Sara, Botti, Silvana, Curtarolo, Stefano, Draxl, Claudia, Cobas, Luis Edmundo Fuentes, Hospital, Adam, Liu, Zi-Kui, Marques, Miguel A. L., Marzari, Nicola, Morris, Andrew J., Ong, Shyue Ping, Orozco, Modesto, Persson, Kristin A., Thygesen, Kristian S., Wolverton, Chris, Scheidgen, Markus, Toher, Cormac, Conduit, Gareth J., Pizzi, Giovanni, Gražulis, Saulius, Rignanese, Gian-Marco, Armiento, Rickard
Publikováno v:
Digital Discovery, 2024, 3, 1509-1533
The Open Databases Integration for Materials Design (OPTIMADE) application programming interface (API) empowers users with holistic access to a growing federation of databases, enhancing the accessibility and discoverability of materials and chemical
Externí odkaz:
http://arxiv.org/abs/2402.00572
Publikováno v:
Proceedings of the 36th Conference on Neural Information Processing Systems (NeurIPS 2022)
We present coarse-to-fine autoregressive networks (C2FAR), a method for modeling the probability distribution of univariate, numeric random variables. C2FAR generates a hierarchical, coarse-to-fine discretization of a variable autoregressively; progr
Externí odkaz:
http://arxiv.org/abs/2312.15002
Publikováno v:
Proceedings of 37th Conference on Neural Information Processing Systems (NeurIPS 2023)
We propose SutraNets, a novel method for neural probabilistic forecasting of long-sequence time series. SutraNets use an autoregressive generative model to factorize the likelihood of long sequences into products of conditional probabilities. When ge
Externí odkaz:
http://arxiv.org/abs/2312.14880
Deep learning models have become widely adopted in various domains, but their performance heavily relies on a vast amount of data. Datasets often contain a large number of irrelevant or redundant samples, which can lead to computational inefficiencie
Externí odkaz:
http://arxiv.org/abs/2309.11922
Autor:
Bergsma, Fenna, author
Publikováno v:
The Place of Case in Grammar, 2024.
Externí odkaz:
https://doi.org/10.1093/oso/9780198865926.003.0013
Autor:
Ishida, Sahoko, Bergsma, Wicher
This paper focuses on interpretable additive Gaussian process (GP) regression and its efficient implementation for large-scale data with a multi-dimensional grid structure, as commonly encountered in spatio-temporal analysis. A popular and scalable a
Externí odkaz:
http://arxiv.org/abs/2305.07073
Autor:
Rudas, Tamas, Bergsma, Wicher
Marginal models involve restrictions on the conditional and marginal association structure of a set of categorical variables. They generalize log-linear models for contingency tables, which are the fundamental tools for modelling the conditional asso
Externí odkaz:
http://arxiv.org/abs/2304.03380
Autor:
Caio M. Massari, Dylan J. Dues, Alexis Bergsma, Kayla Sipple, Maxwell Frye, Erin T. Williams, Darren J. Moore
Publikováno v:
Neurobiology of Disease, Vol 202, Iss , Pp 106701- (2024)
Loss-of-function mutations in the ATP13A2 (PARK9) gene are implicated in early-onset autosomal recessive Parkinson's disease (PD) and other neurodegenerative disorders. ATP13A2 encodes a lysosomal transmembrane P5B-type ATPase that is highly expresse
Externí odkaz:
https://doaj.org/article/164500fdad7f44058d1638080cd23172