| Description |
1 online resource (x, 233 pages) : illustrations |
| Series |
Communications in computer and information science, 1865-0929 ; 908 |
|
Communications in computer and information science ; 908. 1865-0929
|
| Contents |
Intro -- Preface -- Organization -- Contents -- Accelerators -- A Scalable FPGA Accelerator for Convolutional Neural Networks -- 1 Introduction -- 2 Background -- 2.1 Classic Convolution Neural Network -- 2.2 OpenCL Framework on FPGA -- 3 Architecture Design and Optimization -- 3.1 Accelerator Architecture -- 4 Design Space Exploration -- 4.1 Performance Model -- 4.2 Resource Utilization Model -- 5 Experimental Results -- 5.1 Experimental Setup -- 5.2 Results and Discussion -- 6 Conclusion -- References |
|
Memory Bandwidth and Energy Efficiency Optimization of Deep Convolutional Neural Network Accelerators -- 1 Introduction -- 2 Background and Related Works -- 2.1 Convolutional Neural Networks -- 2.2 Related Works -- 3 Motivation -- 3.1 Limited On-Chip Storage Resources -- 3.2 Data Movement -- 4 Data Layout Optimization -- 4.1 Computation and Data Layout -- 4.2 NHWCcoarse: A Coarse-grained Optimization of Computation Sequence -- 4.3 NvHWCfine: A Fine-grained Optimization of Computation Sequence -- 4.4 Segmentation -- 4.5 Parallelization -- 5 Experiment Setup -- 5.1 Benchmark and Baseline |
|
5.2 Experiment Framework -- 6 Experiment Results -- 6.1 The Number of Off-Chip DRAM Accesses -- 6.2 Total Latency of Off-Chip DRAM Accesses -- 6.3 Energy Consumption -- 7 Conclusions and Future Works -- References -- Research on Parallel Acceleration for Deep Learning Inference Based on Many-Core ARM Platform -- Abstract -- 1 Introduction -- 2 The Many-Core ARM -- 3 Parallel Resources of ARM -- 4 Parallel Computing Library of ARM -- 5 Target Applications -- 5.1 RNNs-LSTM Model -- 5.2 Fully Connected Model -- 5.3 Confusion Tree Model -- 6 The Design of Experiments |
|
6.1 Settings of Platform Comparison -- 6.2 Parallel Granularity -- 6.3 Parallel Scheme for RNNs-LSTM -- 6.4 Parallel Scheme for Full Connected Model -- 6.5 Parallel Scheme for Confusion Tree Model -- 6.6 Optimization Technology -- 7 Experimental Results -- 7.1 Results of RNNs-LSTM Model -- 7.2 Results of Full Connected Model -- 7.3 Results of Confusion Tree Model -- 7.4 Energy Efficiency Comparison -- 8 Conclusion -- References -- Research on Acceleration Method of Speech Recognition Training -- Abstract -- 1 Introduction -- 2 Methods -- 2.1 Acoustic Characteristics -- 2.2 Network Model |
|
2.2.1 RNN Network -- 2.2.2 Speech Recognition Architecture -- 2.3 Text Correction -- 3 Experiments -- 3.1 Result -- 4 Conclusion and Future Work -- References -- New Design Explorations -- A Post-link Prefetching Based on Event Sampling -- Abstract -- 1 Introduction -- 2 Related Work -- 3 Post-link Prefetching Based on Event Sampling -- 3.1 Event Based Sampling -- 3.2 Sample Analysis -- 3.3 Inserting Prefetch Instructions -- 4 Experiment Settings -- 5 Evaluation -- 5.1 Performance Evaluation of the Optimized Prefetching -- 5.2 The Evaluation of the Sampling Overhead -- 6 Conclusion |
| Summary |
This book constitutes the refereed proceedings of the 12th Annual Conference on Advanced Computer Architecture, ACA 2018, held in Yingkou, China, in August 2018. The 17 revised full papers presented were carefully reviewed and selected from 80 submissions. The papers of this volume are organized in topical sections on: accelerators; new design explorations; towards efficient ML/AI; parallel computing system |
| Notes |
International conference proceedings |
|
Includes author index |
|
Online resource; title from PDF title page (SpringerLink, viewed September 24, 2018) |
| Subject |
Computer architecture -- Congresses
|
|
Expert systems -- knowledge-based systems.
|
|
Systems analysis & design.
|
|
Operating systems.
|
|
Computer architecture & logic design.
|
|
Computers -- Expert Systems.
|
|
Computers -- Information Technology.
|
|
Computers -- Operating Systems -- General.
|
|
Computers -- Logic Design.
|
|
Computers -- Systems Architecture -- General.
|
|
Computer architecture
|
| Genre/Form |
proceedings (reports)
|
|
Conference papers and proceedings
|
|
Conference papers and proceedings.
|
|
Actes de congrès.
|
| Form |
Electronic book
|
| Author |
Li, Chao, editor
|
|
Wu, Junjie (College teacher), editor.
|
| ISBN |
9789811324239 |
|
9811324239 |
|
9789811324246 |
|
9811324247 |
|
