Michael J. Houston
University of Houston
Valence (psychology)Deep learningAdvertisingProduct (category theory)SociologyBusinessArtificial intelligencePsychologyMarketingCross-culturalCognitionCognitive psychologyInterdependenceChinaConsumer behaviourValue (ethics)Context (language use)Meaning (existential)Computer scienceArtificial neural networkMeaning (linguistics)Brand managementBrand extensionSocial psychologyInformation processing
118Publications
42H-index
8,839Citations
Publications 64
Newest
#1Liu Yang (Brown University)H-Index: 8
#2Prabhat (LBNL: Lawrence Berkeley National Laboratory)H-Index: 31
Last. Michael J. Houston (Nvidia)H-Index: 42
view all 11 authors...
Uncertainty quantification for forward and inverse problems is a central challenge across physical and biomedical disciplines. We address this challenge for the problem of modeling subsurface flow at the Hanford Site by combining stochastic computational models with observational data using physics-informed GAN models. The geographic extent, spatial heterogeneity, and multiple correlation length scales of the Hanford Site require training a computationally intensive GAN model to thousands of dim...
4 CitationsSource
#1Liu YangH-Index: 8
#2Sean J. TreichlerH-Index: 11
Last. George Em KarniadakisH-Index: 110
view all 11 authors...
Uncertainty quantification for forward and inverse problems is a central challenge across physical and biomedical disciplines. We address this challenge for the problem of modeling subsurface flow at the Hanford Site by combining stochastic computational models with observational data using physics-informed GAN models. The geographic extent, spatial heterogeneity, and multiple correlation length scales of the Hanford Site require training a computationally intensive GAN model to thousands of dim...
#1Liu Yang (Brown University)H-Index: 8
#2Sean J. Treichler (Nvidia)H-Index: 11
Last. Prabhat (LBNL: Lawrence Berkeley National Laboratory)H-Index: 31
view all 11 authors...
Uncertainty quantification for forward and inverse problems is a central challenge across physical and biomedical disciplines. We address this challenge for the problem of modeling subsurface flow at the Hanford Site by combining stochastic computational models with observational data using physics-informed GAN models. The geographic extent, spatial heterogeneity, and multiple correlation length scales of the Hanford Site require training a computationally intensive GAN model to thousands of dim...
8 Citations
#1Agnieszka Zablocki (WU: Vienna University of Economics and Business)H-Index: 2
#2Bodo B. Schlegelmilch (WU: Vienna University of Economics and Business)H-Index: 57
Last. Michael J. Houston (UMN: University of Minnesota)H-Index: 42
view all 3 authors...
Online reviews can strongly influence purchase decisions. In the past decade, extensive research in the field of online reviews has focused on product categories (e.g., hedonic, utilitarian) and product sales. However, research on how the characteristics of online reviews (valence, volume, and variance) influence attitudes toward brands is sparse, even though brands are among the most valuable corporate assets and companies use online marketing extensively to increase brand loyalty. Thus, this p...
2 CitationsSource
#1Agnieszka ZablockiH-Index: 2
#2Katerina Makri (WU: Vienna University of Economics and Business)H-Index: 6
Last. Michael J. Houston (UMN: University of Minnesota)H-Index: 42
view all 4 authors...
Source
#1Agnieszka ZablockiH-Index: 2
#2Katerina Makri (WU: Vienna University of Economics and Business)H-Index: 6
Last. Michael J. Houston (UMN: University of Minnesota)H-Index: 42
view all 3 authors...
Abstract Online reviews are of considerable interest to both practitioners and academics. Prior research has focused mostly on the valence and volume of online reviews. Questions on how the emotional content of online reviews influence consumers remain open. Given that non-verbal cues of communication are limited in the online environment, content can be a strong driver of attitudes and, in turn, online purchases. Importantly also, a wide stream of research indicates that different self-construa...
8 CitationsSource
#1Prabhat (LBNL: Lawrence Berkeley National Laboratory)H-Index: 31
#2Thorsten Kurth (LBNL: Lawrence Berkeley National Laboratory)H-Index: 26
Last. William D. Collins (UC: University of California)H-Index: 56
view all 14 authors...
#1M. PrabhatH-Index: 2
#2Thorsten KurthH-Index: 2
Last. William D. CollinsH-Index: 56
view all 14 authors...
#1Thorsten Kurth (LBNL: Lawrence Berkeley National Laboratory)H-Index: 26
#2Sean J. Treichler (Nvidia)H-Index: 11
Last. Michael J. Houston (Nvidia)H-Index: 42
view all 12 authors...
We extract pixel-level masks of extreme weather patterns using variants of Tiramisu and DeepLabv3+ neural networks. We describe improvements to the software frameworks, input pipeline, and the network training algorithms necessary to efficiently scale deep learning on the Piz Daint and Summit systems. The Tiramisu network scales to 5300 P100 GPUs with a sustained throughput of 21.0 PF/s and parallel efficiency of 79.0%. DeepLabv3+ scales up to 27360 V100 GPUs with a sustained throughput of 325.8...
70 CitationsSource
We extract pixel-level masks of extreme weather patterns using variants of Tiramisu and DeepLabv3+ neural networks. We describe improvements to the software frameworks, input pipeline, and the network training algorithms necessary to efficiently scale deep learning on the Piz Daint and Summit systems. The Tiramisu network scales to 5300 P100 GPUs with a sustained throughput of 21.0 PF/s and parallel efficiency of 79.0%. DeepLabv3+ scales up to 27360 V100 GPUs with a sustained throughput of 325.8...
45 Citations