75,763 Hits in 4.6 sec

Neural methods for dynamic branch prediction

Daniel A. Jiménez, Calvin Lin
2002 ACM Transactions on Computer Systems  
This article presents a new and highly accurate method for branch prediction.  ...  The key idea is to use one of the simplest possible neural methods, the perceptron, as an alternative to the commonly used two-bit counters.  ...  ACKNOWLEDGMENTS We thank the anonymous referees for their valuable comments.  ... 
doi:10.1145/571637.571639 fatcat:qfztavkunfbrbkjwxvjvhng3je

Recursive-NeRF: An Efficient and Dynamically Growing NeRF [article]

Guo-Wei Yang, Wen-Yang Zhou, Hao-Yang Peng, Dun Liang, Tai-Jiang Mu, Shi-Min Hu
2021 arXiv   pre-print
The core of Recursive-NeRF learns uncertainties for query coordinates, representing the quality of the predicted color and volumetric intensity at each level.  ...  View synthesis methods using implicit continuous shape representations learned from a set of images, such as the Neural Radiance Field (NeRF) method, have gained increasing attention due to their high  ...  We would like to thank Guo-Ye Yang for his kindly help in experimentation and Prof. Ralph R. Martin for his help in writing.  ... 
arXiv:2105.09103v1 fatcat:zwm6shf425f5tc5nq5sxb5ywm4

A Deep Learning and GIS Approach for the Optimal Positioning of Wave Energy Converters

Georgios Batsis, Panagiotis Partsinevelos, Georgios Stavrakakis
2021 Energies  
The branch that is trained with image data provides for the localization of dynamic geospatial classes in the potential installation area, whereas the second one is responsible for the classification of  ...  These data are combined with neural network predictions for the optimizing positioning of the Wave Energy Converters.  ...  Data Fusion Based Neural Network Consequently, we developed a Convolutional architecture for each of two branches.  ... 
doi:10.3390/en14206773 fatcat:24hrnet5sreoflniirnkndnpn4

A Dynamic branch predictor Based on Parallel structure of SRNN

Lei Zhang, Ning Wu, Fen Ge, Fang Zhou, Mahammad Rehan Yahya
2020 IEEE Access  
, and its branch prediction rate is 2.34% higher than the traditional Perceptron neural predictor in the short learning period.  ...  In this paper, a dynamic branch predictor based on parallel structure of SRNN is proposed to accelerate the training time and reduces the computing delay.  ...  explored more advanced branch prediction machine learning methods [9] .  ... 
doi:10.1109/access.2020.2992643 fatcat:dvafw7cdx5dgtoafvyqqhjqafa

Improving Accuracy of Perceptron Predictor Through Correlating Data Values in SMT Processors [chapter]

Liqiang He, Zhiyong Liu
2005 Lecture Notes in Computer Science  
Many predictors based on neural network, especially on perceptron, are proposed to provide a more accurate dynamic branch prediction than before in the literature.  ...  The key idea is using a dynamic bias input, which comes from some information independent on the branch histories (data values for example), to realize the objective of improving accuracy.  ...  This static bias input setting is similar as a static branch prediction in the past. It is well known that a static branch prediction scheme has a lower accuracy than the dynamic prediction schemes.  ... 
doi:10.1007/11427469_151 fatcat:vovcf7onqnfhnfr3ttbwaa7xgy

NDT: Neual Decision Tree Towards Fully Functioned Neural Graph [article]

Han Xiao
2017 arXiv   pre-print
Within our novel principle, we propose the neural decision tree (NDT), which takes simplified neural networks as decision function in each branch and employs complex neural networks to generate the output  ...  Though traditional algorithms could be embedded into neural architectures with the proposed principle of xiao2017hungarian, the variables that only occur in the condition of branch could not be updated  ...  To tackle the issue of overfull categories, we propose the method of neural decision tree (NDT), which takes simplified neural networks as decision function in each branch and employs complex neural networks  ... 
arXiv:1712.05934v1 fatcat:qc7lqp7yivdj7acheihgqkuetq

An astute LVQ approach using neural network for the prediction of conditional branches in pipeline processor

Sweety Nain, Prachi Chaudhary
2018 EAI Endorsed Transactions on Scalable Information Systems  
Firstly, the perceptron neural network and global-based perceptron prediction has been exploited and implemented.  ...  However, recent processors still have problems with the correct prediction of conditional branches.  ...  Prachi Chaudhary for providing many helpful contributions during this paper.  ... 
doi:10.4108/eai.4-3-2021.168865 fatcat:f3jxawxyx5fjpmypzc2zd5237m

Two-level branch prediction using neural networks

Colin Egan, Gordon Steven, Patrick Quick, Rubén Anguera, Fleur Steven, Lucian Vintan
2003 Journal of systems architecture  
An alternative approach is to look to other application areas and fields for novel solutions to the problem. In this paper, we examine the application of neural networks to dynamic branch prediction.  ...  Dynamic branch prediction in high-performance processors is a specific instance of a general time series prediction problem that occurs in many areas of science.  ...  Jim e enez, independently and simultaneously to our investigation into the potential of applying the principles of NNs to dynamic branch prediction, has also investigated neural methods for dynamic branch  ... 
doi:10.1016/s1383-7621(03)00095-x fatcat:hqbzt2on2zfstlni4ij7osxnqy

Learning Execution through Neural Code Fusion [article]

Zhan Shi, Kevin Swersky, Daniel Tarlow, Parthasarathy Ranganathan, Milad Hashemi
2020 arXiv   pre-print
As an illustration of this, we apply the new model to challenging dynamic tasks (branch prediction and prefetching) from the SPEC CPU benchmark suite, outperforming the state-of-the-art by 26% and 45%  ...  While there is a growing body of work on using Graph Neural Networks (GNNs) to learn representations of source code, these representations do not understand how code dynamically executes.  ...  DYNAMIC PREDICTION TASKS Branch prediction and prefetching are heavily studied in the computer architecture domain.  ... 
arXiv:1906.07181v2 fatcat:czxoxlbifbgn3kvssezsyzjclu

Balancing the learning ability and memory demand of a perceptron-based dynamically trainable neural network

Edward Richter, Spencer Valancius, Josiah McClanahan, John Mixter, Ali Akoglu
2018 Journal of Supercomputing  
We use the well-known perceptron-based branch prediction problem as a case study for demonstrating this methodology.  ...  Two challenges prominent in the neural network domain are the practicality of hardware implementation and dynamically training the network.  ...  Jimenez and Lin [13] proposed a dynamic branch prediction scheme using neural networks to predict branches at run time.  ... 
doi:10.1007/s11227-018-2374-x fatcat:nuirrf47erc4lamcbiby7ul75m

Gated Residual Recurrent Graph Neural Networks for Traffic Prediction

Cen Chen, Kenli Li, Sin G. Teo, Xiaofeng Zou, Kang Wang, Jie Wang, Zeng Zeng
are utilized for spatial dependency and RNNs for temporal dynamics.  ...  In the literature, many research works have applied deep learning methods on traffic prediction problems combining convolutional neural networks (CNNs) with recurrent neural networks (RNNs), which CNNs  ...  not utilized, (iv) RGNN denotes the proposed recurrent graph neural networks with only a branch for the near temporal dependency without gated residual shortcuts, (v) Res-RGNN-G stands for one branch of  ... 
doi:10.1609/aaai.v33i01.3301485 fatcat:wvazfis5arcvhm3dwdwadftari

DmifNet:3D Shape Reconstruction Based on Dynamic Multi-Branch Information Fusion [article]

Lei Li, Suping Wu
2020 arXiv   pre-print
Finally, we dynamically fuse the information of all branches to gain final predicted probability.  ...  In this paper, we propose a Dynamic Multi-branch Information Fusion Network (DmifNet) which can recover a high-fidelity 3D shape of arbitrary topology from a 2D image.  ...  Finally, we dynamically fuse the prediction results of the main branch and the side branches.  ... 
arXiv:2011.10776v1 fatcat:m6dpww5f6zh2xaplqooz6quwca

DeepGraphONet: A Deep Graph Operator Network to Learn and Zero-shot Transfer the Dynamic Response of Networked Systems [article]

Yixuan Sun, Christian Moya, Guang Lin, Meng Yue
2022 arXiv   pre-print
The resulting DeepGraphONet can then predict the dynamics within a given short/medium-term time horizon by observing a finite history of the graph state information.  ...  operator of dynamical systems.  ...  In [9] , the authors proposed a multistep method with a feed-forward neural network to approximate the dynamical system response.  ... 
arXiv:2209.10622v1 fatcat:ogtqu24kufachdx64bu3o6amz4

Exploring Convolution Neural Network for Branch Prediction

Yonghua Mao, Huiyang Zhou, Xiaolin Gui, Junjie Shen
2020 IEEE Access  
However, only a few works [22] explored more advanced machine learning methods on the branch prediction problem, much less deep neural networks.  ...  We explore both deep belief network (DBN) and convolutional neural networks (CNNs) for branch prediction.  ... 
doi:10.1109/access.2020.3017196 fatcat:hp3we2bp4zfjjbcxv3a7bwdb4y

Research on Short-term Traffic Flow Forecasting for Junction of Isomerism Road Network based on Dynamic Correlation

Liang-liang Zhang, Yuan-hua Jia, Zhong-hai Niu, Hua-nan Li
2014 Procedia - Social and Behavioral Sciences  
Following, the paper selects input variables, and establishes the Radial Basic Function (RBF) neural network model for prediction on the basis of dynamic correlation coefficient.  ...  There are various methods that have been established to forecast traffic flow, but most of the forecasting models are constructed according to analysis of the historical and current traffic flow series  ...  Acknowledgements Specially thanks to research subject that is Research on Traffic Cooperative Control Theory and Method for the Junction of Isomerism Road Network System (71340020), supported by the National  ... 
doi:10.1016/j.sbspro.2014.07.223 fatcat:m7xqj4snejhwpmjnsvmmigaozu
« Previous Showing results 1 — 15 out of 75,763 results