Using deep learning for precipitation forecasting based on spatio-temporal information: a case study

2021 ◽  
Author(s):  
Weide Li ◽  
Xi Gao ◽  
Zihan Hao ◽  
Rong Sun
Keyword(s):  
Deep Learning ◽  
Temporal Information ◽  
Precipitation Forecasting ◽  
Spatio Temporal

Deep learning-based hierarchical cattle behavior recognition with spatio-temporal information

2020 ◽  
Vol 177 ◽  
pp. 105627
Author(s):  
Alvaro Fuentes ◽  
Sook Yoon ◽  
Jongbin Park ◽  
Dong Sun Park
Keyword(s):  
Deep Learning ◽  
Temporal Information ◽  
Behavior Recognition ◽  
Spatio Temporal

scholarly journals Deep Learning Accelerators: A Case Study with MAESTRO

2020 ◽  
Author(s):  
Hamidreza Bolhasani ◽  
Somayyeh Jafarali Jassbi
Keyword(s):  
Deep Learning ◽  
Language Processing ◽  
Performance Indicators ◽  
Learning Task ◽  
High Accuracy ◽  
Spatio Temporal ◽  
Almost All ◽  
L1 And L2 ◽  
Accuracy Speed

Abstract In recent years, deep learning has become one of the most important topics in computer sciences. Deep learning is a growing trend in the edge of technology and its applications are now seen in many aspects of our life such as object detection, speech recognition, natural language processing, etc. Currently, almost all major sciences and technologies are benefiting from the advantages of deep learning such as high accuracy, speed and flexibility. Therefore, any efforts in improving performance of related techniques is valuable. Deep learning accelerators are considered as hardware architecture, which are designed and optimized for increasing speed, efficiency and accuracy of computers that are running deep learning algorithms. In this paper, after reviewing some backgrounds on deep learning, a well-known accelerator architecture named MAERI (Multiply-Accumulate Engine with Reconfigurable interconnects) is investigated. Performance of a deep learning task is measured and compared in two different data flow strategies: NLR (No Local Reuse) and NVDLA (NVIDIA Deep Learning Accelerator), using an open source tool called MAESTRO (Modeling Accelerator Efficiency via Spatio-Temporal Resource Occupancy). Measured performance indicators of novel optimized architecture, NVDLA shows higher L1 and L2 computation reuse, and lower total runtime (cycles) in comparison to the other one.

scholarly journals Mapping historical landscape changes with the use of a space-time cube

2018 ◽  
Vol 1 ◽  
pp. 1-3
Author(s):  
Edyta P. Bogucka ◽  
Mathias Jahnke
Keyword(s):  
Landscape Structure ◽  
Space Time ◽  
Temporal Information ◽  
Landscape Changes ◽  
Map Projections ◽  
Historical Landscape ◽  
Spatio Temporal

In this contribution, we introduce geographic concepts in the humanities and present the results of a spacetime visualization of ancient buildings over the last centuries. The techniques and approaches used were based on cartographic research to visualize spatio-temporal information. As a case study, we applied cartographic styling techniques to a model of the Royal Castle in Warsaw and its different spatial elements, which were constructed and destroyed during their eventful history. In our case, the space-time cube approach seems to be the most suitable representation of this spatio-temporal information. Therefore, we digitized the different footprints of the castle during the ancient centuries as well as the landscape structure around, and annotated them with monarchies, epochs and time. During the digitization process, we had to cope with difficulties like sources in various scales and map projections, which resulted in varying accuracies. The results were stored in KML to support a wide variety of visualization platforms.

scholarly journals Deep Learning Accelerators: A Case Study with MAESTRO

2020 ◽  
Author(s):  
Hamidreza Bolhasani ◽  
Somayyeh Jafarali Jassbi
Keyword(s):  
Deep Learning ◽  
Language Processing ◽  
Performance Indicators ◽  
Learning Task ◽  
High Accuracy ◽  
Spatio Temporal ◽  
Almost All ◽  
L1 And L2 ◽  
Accuracy Speed

Abstract In recent years, deep learning has become one of the most important topics in computer sciences. Deep learning is a growing trend in the edge of technology and its applications are now seen in many aspects of our life such as object detection, speech recognition, natural language processing, etc. Currently, almost all major sciences and technologies are benefiting from the advantages of deep learning such as high accuracy, speed and flexibility. Therefore, any efforts in improving performance of related techniques is valuable. Deep learning accelerators are considered as hardware architecture, which are designed and optimized for increasing speed, efficiency and accuracy of computers that are running deep learning algorithms. In this paper, after reviewing some backgrounds on deep learning, a well-known accelerator architecture named MAERI (Multiply-Accumulate Engine with Reconfigurable interconnects) is investigated. Performance of a deep learning task is measured and compared in two different data flow strategies: NLR (No Local Reuse) and NVDLA (NVIDIA Deep Learning Accelerator), using an open source tool called MAESTRO (Modeling Accelerator Efficiency via Spatio-Temporal Resource Occupancy). Measured performance indicators of novel optimized architecture, NVDLA shows higher L1 and L2 computation reuse, and lower total runtime (cycles) in comparison to the other one.

scholarly journals Deep learning accelerators: a case study with MAESTRO

2020 ◽  
Vol 7 (1) ◽  
Author(s):  
Hamidreza Bolhasani ◽  
Somayyeh Jafarali Jassbi
Keyword(s):  
Deep Learning ◽  
Language Processing ◽  
Performance Indicators ◽  
Learning Task ◽  
High Accuracy ◽  
Spatio Temporal ◽  
Almost All ◽  
L1 And L2 ◽  
Accuracy Speed

AbstractIn recent years, deep learning has become one of the most important topics in computer sciences. Deep learning is a growing trend in the edge of technology and its applications are now seen in many aspects of our life such as object detection, speech recognition, natural language processing, etc. Currently, almost all major sciences and technologies are benefiting from the advantages of deep learning such as high accuracy, speed and flexibility. Therefore, any efforts in improving performance of related techniques is valuable. Deep learning accelerators are considered as hardware architecture, which are designed and optimized for increasing speed, efficiency and accuracy of computers that are running deep learning algorithms. In this paper, after reviewing some backgrounds on deep learning, a well-known accelerator architecture named MAERI (Multiply-Accumulate Engine with Reconfigurable interconnects) is investigated. Performance of a deep learning task is measured and compared in two different data flow strategies: NLR (No Local Reuse) and NVDLA (NVIDIA Deep Learning Accelerator), using an open source tool called MAESTRO (Modeling Accelerator Efficiency via Spatio-Temporal Resource Occupancy). Measured performance indicators of novel optimized architecture, NVDLA shows higher L1 and L2 computation reuse, and lower total runtime (cycles) in comparison to the other one.

Spatio-temporal approach for identification of critical conservation areas: a case study with two pine species from a threatened temperate forest in Mexico

2019 ◽  
Vol 28 (7) ◽  
pp. 1863-1883 ◽  
Author(s):  
Agustín Molina Sánchez ◽  
Patricia Delgado ◽  
Antonio González-Rodríguez ◽  
Clementina González ◽  
A. Francisco Gómez-Tagle Rojas ◽  
...  
Keyword(s):  
Temperate Forest ◽  
Conservation Areas ◽  
Spatio Temporal ◽  
Pine Species

scholarly journals A comparison of multiple neighborhood matrix specifications for spatio-temporal model fitting: a case study on COVID-19 data

2021 ◽  
Author(s):  
Álvaro Briz-Redón ◽  
Adina Iftimi ◽  
Juan Francisco Correcher ◽  
Jose De Andrés ◽  
Manuel Lozano ◽  
...  
Keyword(s):  
Model Fitting ◽  
Temporal Model ◽  
Spatio Temporal

scholarly journals Spatio-temporal analysis of COVID-19 incidence rate using GIS: a case study—Tehran metropolitan, Iran

GeoJournal ◽  
2021 ◽  
Author(s):  
R. Nasiri ◽  
S. Akbarpour ◽  
AR. Zali ◽  
N. Khodakarami ◽  
MH. Boochani ◽  
...  
Keyword(s):  
Incidence Rate ◽  
Temporal Analysis ◽  
Spatio Temporal

scholarly journals Applications of deep learning to decorated ceramic typology and classification: A case study using Tusayan White Ware from Northeast Arizona

2021 ◽  
Vol 130 ◽  
pp. 105375
Author(s):  
Leszek M. Pawlowicz ◽  
Christian E. Downum
Keyword(s):  
Deep Learning ◽  
Ceramic Typology
Sign in / Sign up

Export Citation Format

Share Document