scholarly journals Crafting Efficient Neural Graph of Large Entropy

2019 ◽  
Author(s):  
Minjing Dong ◽  
Hanting Chen ◽  
Yunhe Wang ◽  
Chang Xu
Keyword(s):  
Information Flow ◽  
Network Architecture ◽  
High Performance ◽  
Global Information ◽  
High Quality ◽  
Network Pruning ◽  
Initial Network ◽  
Trade Offs ◽  
Graph Properties ◽  
Deep Cnn

Network pruning is widely applied to deep CNN models due to their heavy computation costs and achieves high performance by keeping important weights while removing the redundancy. Pruning redundant weights directly may hurt global information flow, which suggests that an efficient sparse network should take graph properties into account. Thus, instead of paying more attention to preserving important weight, we focus on the pruned architecture itself. We propose to use graph entropy as the measurement, which shows useful properties to craft high-quality neural graphs and enables us to propose efficient algorithm to construct them as the initial network architecture. Our algorithm can be easily implemented and deployed to different popular CNN models and achieve better trade-offs.

You Only Traverse Twice: A YOTT Placement, Routing, and Timing Approach for CGRAs

2021 ◽  
Vol 20 (5s) ◽  
pp. 1-25
Author(s):  
Michael Canesche ◽  
Westerley Carvalho ◽  
Lucas Reis ◽  
Matheus Oliveira ◽  
Salles Magalhães ◽  
...  
Keyword(s):  
Execution Time ◽  
High Performance ◽  
Coarse Grained ◽  
Optimal Placement ◽  
Greedy Heuristics ◽  
High Quality ◽  
Solution Quality ◽  
Graph Traversal ◽  
Trade Offs ◽  
Graph Properties

Coarse-grained reconfigurable architecture (CGRA) mapping involves three main steps: placement, routing, and timing. The mapping is an NP-complete problem, and a common strategy is to decouple this process into its independent steps. This work focuses on the placement step, and its aim is to propose a technique that is both reasonably fast and leads to high-performance solutions. Furthermore, a near-optimal placement simplifies the following routing and timing steps. Exact solutions cannot find placements in a reasonable execution time as input designs increase in size. Heuristic solutions include meta-heuristics, such as Simulated Annealing (SA) and fast and straightforward greedy heuristics based on graph traversal. However, as these approaches are probabilistic and have a large design space, it is not easy to provide both run-time efficiency and good solution quality. We propose a graph traversal heuristic that provides the best of both: high-quality placements similar to SA and the execution time of graph traversal approaches. Our placement introduces novel ideas based on “you only traverse twice” (YOTT) approach that performs a two-step graph traversal. The first traversal generates annotated data to guide the second step, which greedily performs the placement, node per node, aided by the annotated data and target architecture constraints. We introduce three new concepts to implement this technique: I/O and reconvergence annotation, degree matching, and look-ahead placement. Our analysis of this approach explores the placement execution time/quality trade-offs. We point out insights on how to analyze graph properties during dataflow mapping. Our results show that YOTT is 60.6 , 9.7 , and 2.3 faster than a high-quality SA, bounding box SA VPR, and multi-single traversal placements, respectively. Furthermore, YOTT reduces the average wire length and the maximal FIFO size (additional timing requirement on CGRAs) to avoid delay mismatches in fully pipelined architectures.

scholarly journals Biopolymer Networks: Image Analysis, Reconstruction and Modeling

2019 ◽  
Author(s):  
Pablo Hernandez-Cerdan
Keyword(s):  
Image Analysis ◽  
Network Architecture ◽  
High Performance ◽  
Network Formation ◽  
Network Size ◽  
Spatial Graphs ◽  
Transmission Electron ◽  
Graph Properties ◽  
Computational Exploration ◽  
Analytical Tools

The aim of this work is to extract the architecture of biopolymer networks from 3D images. This is motivated to further understand non-affine regimes, found during network formation and in low-density biopolymer networks, where the geometry of a network has a fundamental role in defining its mechanical properties. Firstly, developed image analysis tools were extended to 3D and contributed to high-performance open-source libraries for image analysis. These developments in isotropic wavelets will help in extracting realistic networks by removing spurious noise generated during image acquisition. Secondly, images of biopolymer gels from transmission electron microscopy (TEM), were used to reliably extract the network architecture. The imaged material was also studied with small-angle x-ray scattering (SAXS), and the comparison showed a strong agreement for network-size features. Thirdly, spatial graphs were extracted from the image. A one-to-one map is provided between image and graph, keeping all the geometric information from the image. This then opened the door to using analytical tools from the complex networks field to characterize images. Finally, statistical distributions extracted from three graph properties were used to reconstruct a completely in-silico network using a simulated annealing technique to generate new networks. This can be used as a computational exploration tool of how network behavior depends on network architecture.

scholarly journals Flux Method Growth of Large Size Group IV–V 2D GeP Single Crystals and Photoresponse Application

Crystals ◽  
2021 ◽  
Vol 11 (3) ◽  
pp. 235
Author(s):  
Shuqi Zhao ◽  
Tongtong Yu ◽  
Ziming Wang ◽  
Shilei Wang ◽  
Limei Wei ◽  
...  
Keyword(s):  
Raman Spectroscopy ◽  
Single Crystals ◽  
High Performance ◽  
Group Iv ◽  
X Ray Diffraction ◽  
High Quality ◽  
Practical Applications ◽  
Large Size ◽  
Growth Method ◽  
Polarized Raman

Two-dimensional (2D) materials driven by their unique electronic and optoelectronic properties have opened up possibilities for their various applications. The large and high-quality single crystals are essential to fabricate high-performance 2D devices for practical applications. Herein, IV-V 2D GeP single crystals with high-quality and large size of 20 × 15 × 5 mm3 were successfully grown by the Bi flux growth method. The crystalline quality of GeP was confirmed by high-resolution X-ray diffraction (HRXRD), Laue diffraction, electron probe microanalysis (EPMA) and Raman spectroscopy. Additionally, intrinsic anisotropic optical properties were investigated by angle-resolved polarized Raman spectroscopy (ARPRS) and transmission spectra in detail. Furthermore, we fabricated high-performance photodetectors based on GeP, presenting a relatively large photocurrent over 3 mA. More generally, our results will significantly contribute the GeP crystal to the wide optoelectronic applications.

scholarly journals High performance self-powered photodetector based on 1D Te-2D WS2 mixed-dimensional heterostructure

2021 ◽  
Author(s):  
Lixiang Han ◽  
Mengmeng Yang ◽  
Peiting Wen ◽  
Wei Gao ◽  
nengjie huo ◽  
...  
Keyword(s):  
High Performance ◽  
Van Der Waals ◽  
Sharp Interface ◽  
Two Dimensional ◽  
High Quality ◽  
One Dimensional ◽  
Self Powered

One dimensional (1D)-two dimensional (2D) van der Waals (vdWs) mixed-dimensional heterostructures with advantages of atomically sharp interface, high quality and good compatibility have attracted tremendous attention in recent years. The...

High-Performance Image Filters via Sparse Approximations

2020 ◽  
Vol 3 (2) ◽  
pp. 1-19
Author(s):  
Kersten Schuster ◽  
Philip Trettner ◽  
Leif Kobbelt
Keyword(s):  
High Performance ◽  
Hardware Acceleration ◽  
Optimization Method ◽  
Translation Invariant ◽  
Approximation Quality ◽  
Trade Offs ◽  
Sparse Approximations ◽  
Image Filters ◽  
Good Trade ◽  
And Performance

We present a numerical optimization method to find highly efficient (sparse) approximations for convolutional image filters. Using a modified parallel tempering approach, we solve a constrained optimization that maximizes approximation quality while strictly staying within a user-prescribed performance budget. The results are multi-pass filters where each pass computes a weighted sum of bilinearly interpolated sparse image samples, exploiting hardware acceleration on the GPU. We systematically decompose the target filter into a series of sparse convolutions, trying to find good trade-offs between approximation quality and performance. Since our sparse filters are linear and translation-invariant, they do not exhibit the aliasing and temporal coherence issues that often appear in filters working on image pyramids. We show several applications, ranging from simple Gaussian or box blurs to the emulation of sophisticated Bokeh effects with user-provided masks. Our filters achieve high performance as well as high quality, often providing significant speed-up at acceptable quality even for separable filters. The optimized filters can be baked into shaders and used as a drop-in replacement for filtering tasks in image processing or rendering pipelines.

scholarly journals Semiconductor Heteroepitaxy

Crystals ◽  
2021 ◽  
Vol 11 (3) ◽  
pp. 229
Author(s):  
Roberto Bergamaschini ◽  
Elisa Vitiello
Keyword(s):  
High Performance ◽  
Next Generation ◽  
High Quality

The quest for high-performance and scalable devices required for next-generation semiconductor applications inevitably passes through the fabrication of high-quality materials and complex designs [...]

Nickel sulfide nanoworm network architecture as a binder-free high-performance non-enzymatic glucose sensor

2021 ◽  
Vol 188 (2) ◽  
Author(s):  
Alan Meng ◽  
Xiaocheng Hong ◽  
Haiqin Zhang ◽  
Wenli Tian ◽  
Zhenjiang Li ◽  
...  
Keyword(s):  
Network Architecture ◽  
High Performance ◽  
Nickel Sulfide ◽  
Glucose Sensor ◽  
Binder Free

High-performance inverted planar perovskite solar cells without a hole transport layer via a solution process under ambient conditions

2015 ◽  
Vol 3 (38) ◽  
pp. 19294-19298 ◽  
Author(s):  
Xichang Bao ◽  
Qianqian Zhu ◽  
Meng Qiu ◽  
Ailing Yang ◽  
Yujin Wang ◽  
...  
Keyword(s):  
Relative Humidity ◽  
High Performance ◽  
Perovskite Solar Cells ◽  
Solution Process ◽  
Hole Transport ◽  
Transport Layer ◽  
Hole Transport Layer ◽  
Ambient Conditions ◽  
High Quality ◽  
Ito Glass

High-quality CH3NH3PbI3 perovskite films were directly prepared on simple treated ITO glass in air under a relative humidity of lower than 30%.

scholarly journals Effective ion pathways and 3D conductive carbon networks in bentonite host enable stable and high-rate lithium–sulfur batteries

2021 ◽  
Vol 10 (1) ◽  
pp. 20-33
Author(s):  
Lian Wu ◽  
Yongqiang Dai ◽  
Wei Zeng ◽  
Jintao Huang ◽  
Bing Liao ◽  
...  
Keyword(s):  
Network Architecture ◽  
High Performance ◽  
Lithium Ion ◽  
Reversible Capacity ◽  
High Rate ◽  
Lithium Sulfur Batteries ◽  
Carbon Networks ◽  
Carbon Coated ◽  
Lithium Sulfur ◽  
Initial Capacity

Abstract Fast charge transfer and lithium-ion transport in the electrodes are necessary for high performance Li–S batteries. Herein, a N-doped carbon-coated intercalated-bentonite (Bent@C) with interlamellar ion path and 3D conductive network architecture is designed to improve the performance of Li–S batteries by expediting ion/electron transport in the cathode. The interlamellar ion pathways are constructed through inorganic/organic intercalation of bentonite. The 3D conductive networks consist of N-doped carbon, both in the interlayer and on the surface of the modified bentonite. Benefiting from the unique structure of the Bent@C, the S/Bent@C cathode exhibits a high initial capacity of 1,361 mA h g−1 at 0.2C and achieves a high reversible capacity of 618.1 m Ah g−1 at 2C after 500 cycles with a sulfur loading of 2 mg cm−2. Moreover, with a higher sulfur loading of 3.0 mg cm−2, the cathode still delivers a reversible capacity of 560.2 mA h g−1 at 0.1C after 100 cycles.

Sign in / Sign up

Export Citation Format

Share Document