Filters








5,119 Hits in 2.6 sec

Accelerating combinatorial filter reduction through constraints [article]

Yulin Zhang, Hazhar Rahmani, Dylan A. Shell, Jason M. O'Kane
2020 arXiv   pre-print
But exact combinatorial filter reduction is an NP-complete problem and all current techniques are either inexact or formalized with exponentially many constraints.  ...  Reduction of combinatorial filters involves compressing state representations that robots use. Such optimization arises in automating the construction of minimalist robots.  ...  CONCLUSION This paper accelerates filter minimization through constraints.  ... 
arXiv:2011.03471v1 fatcat:mevttxoxq5dz5nmids2il6rtke

StructADMM: A Systematic, High-Efficiency Framework of Structured Weight Pruning for DNNs [article]

Tianyun Zhang, Shaokai Ye, Kaiqi Zhang, Xiaolong Ma, Ning Liu, Linfeng Zhang, Jian Tang, Kaisheng Ma, Xue Lin, Makan Fardad, Yanzhi Wang
2019 arXiv   pre-print
It is a framework that can be used to induce different types of structured sparsity, such as filter-wise, channel-wise, and shape-wise sparsity, as well non-structured sparsity.  ...  Structured weight pruning methods have been proposed to overcome the limitation of irregular network structure and demonstrated actual GPU acceleration.  ...  However, most reduction is achieved in FC layers, and 2.7× reduction achieved in CONV layers will not lead to an overall acceleration in GPU [Wen et al., 2016] .  ... 
arXiv:1807.11091v3 fatcat:yjjxuxcvmfglvll7iudtlf6jzy

A Novel Multi-thread Parallel Constraint Propagation Scheme

Zhe Li, Zhezhou Yu, Peng Wu, Jianan Chen, Zhanshan Li
2019 IEEE Access  
Constraint Programming (CP) is an efficient technique for solving combinatorial (optimization) problems.  ...  INDEX TERMS Constraint programming, constraint satisfaction problem, parallel constraint propagation, parallel generalized arc consistency, simple tabular reduction.  ...  INTRODUCTION Constraint Programming (CP) is an efficient technique for solving combinatorial (optimization) problems.  ... 
doi:10.1109/access.2019.2951027 fatcat:vlutdmbyqrgl5p72rh7vmyaxme

A Hybridalldifferent-Tabu Search Algorithm for Solving Sudoku Puzzles

Ricardo Soto, Broderick Crawford, Cristian Galleguillos, Fernando Paredes, Enrique Norero
2015 Computational Intelligence and Neuroscience  
Thealldifferentconstraint is known to be efficient for domain filtering in the presence of constraints that must be pairwise different, which are exactly the kind of constraints that Sudokus own.  ...  The Sudoku problem is a well-known logic-based puzzle of combinatorial number-placement.  ...  Figure 2 : 2 Example of alldifferent constraint through Hall's theorem approach.  ... 
doi:10.1155/2015/286354 pmid:26078751 pmcid:PMC4453279 fatcat:wj4yqq2jmvh7ldg557qly7xpzq

Distribution network reconfiguration in smart grid system using modified particle swarm optimization

I. I. Atteya, H. A. Ashour, N. Fahmi, D. Strickland
2016 2016 IEEE International Conference on Renewable Energy Research and Applications (ICRERA)  
DNR Objectives and Constraints DNR is considered highly complex, nonlinear, discrete, combinatorial, stochastic optimization problem [4, 5] .  ...  time by neglecting the infeasible particles based on the current and the voltage constraint after load flow calculations, and this in turn accelerate the algorithm.  ... 
doi:10.1109/icrera.2016.7884556 fatcat:mfdhp3t7tbbnfkecbp7nzuhbjq

SS-Auto: A Single-Shot, Automatic Structured Weight Pruning Framework of DNNs with Ultra-High Efficiency [article]

Zhengang Li, Yifan Gong, Xiaolong Ma, Sijia Liu, Mengshu Sun, Zheng Zhan, Zhenglun Kong, Geng Yuan, Yanzhi Wang
2020 arXiv   pre-print
Structured weight pruning is a representative model compression technique of DNNs for hardware efficiency and inference accelerations.  ...  We adopt soft constraint-based formulation to alleviate the strong non-convexity of l0-norm constraints used in state-of-the-art ADMM-based methods for faster convergence and fewer hyperparameters.  ...  Through avoiding the strongly non-convex 0 -norm based hard constraints in the state-of-the-art ADMM formulation, convergence can be accelerated, and the number of associated hyperparameters can be reduced  ... 
arXiv:2001.08839v1 fatcat:yxw42ag2p5b6xm2bok2pc6qeya

Sliding Basis Optimization for Heterogeneous Material Design [article]

Nurcan Gecer Ulu, Svyatoslav Korneev, Erva Ulu, Saigopal Nelaturi
2020 arXiv   pre-print
This approach allows greater localized control of the material distribution as the sliding window moves through higher frequencies.  ...  The sliding basis framework is independent of any particular physics analysis, objective and constraints, providing a versatile and powerful design optimization tool for various applications.  ...  We utilize logistic function based filters to avoid these extra constraints in both graded and discrete material design.  ... 
arXiv:2005.08838v1 fatcat:emgnf6tnpbgelgrxfv47otmbqq

Discrepancy-Based Additive Bounding Procedures

Andrea Lodi, Michela Milano, Louis-Martin Rousseau
2006 INFORMS journal on computing  
Our key result is the exploitation of the k-discrepancy constraint to improve the bound given by any relaxation of a combinatorial optimization problem through the additive bounding technique (Fischetti  ...  The property that a node has an associated discrepancy k can be modeled (and enforced) through a linear constraint.  ...  A preliminary version of the first part of this paper appeared in the Proceedings of "Principle and Practice of Constraint Programming" . Thanks are due to Matteo Fischetti, Mauro Dell'Amico, and W.  ... 
doi:10.1287/ijoc.1050.0168 fatcat:vtzaml2q3jf5holksddu3bis2m

AutoCompress: An Automatic DNN Structured Pruning Framework for Ultra-High Compression Rates [article]

Ning Liu and Xiaolong Ma and Zhiyuan Xu and Yanzhi Wang and Jian Tang and Jieping Ye
2019 arXiv   pre-print
Structured weight pruning is a representative model compression technique of DNNs to reduce the storage and computation requirements and accelerate inference.  ...  As an example, AutoCompress outperforms the prior work on automatic model compression by up to 33x in pruning rate (120x reduction in the actual parameter count) under the same accuracy.  ...  As a key property, ADMM can effectively deal with a subset of combinatorial constraints and yield optimal (or at least high quality) solutions.  ... 
arXiv:1907.03141v2 fatcat:xojubr4r5jhghn6iw64mjfcium

Combinatorial Interaction Testing of Software Product Lines: A Mapping Study

Mohd Zanes Sahid, Abu Bakar Md Sultan, Abdul Azim Abdul Ghani, Salmi Baharom
2016 Journal of Computer Science  
By following the systematic mapping study guidelines, we have selected and filtered 44 primary studies for review.  ...  This paper reports a Systematic Mapping Study (SMS) of relevant primary studies as the evidence on the application of Combinatorial Interaction Testing (CIT) for SPL.  ...  In 2011, PACOGEN introduced in S29 and proposed a definition of global constraint and filtering algorithm to select only valid test configurations.  ... 
doi:10.3844/jcssp.2016.379.398 fatcat:kdc2r5kkiveqho3ycwcdmmf77u

Peeking in solver strategies using explanations visualization of dynamic graphs for constraint programming

Mohammad Ghoniem, Hadrien Cambazard, Jean-Daniel Fekete, Narendra Jussien
2005 Proceedings of the 2005 ACM symposium on Software visualization - SoftVis '05  
In this paper, we describe the use of visualization tools in the context of constraint programming.  ...  Specifically, we focus on the visualization of dynamic graphs using adjacency matrices, and show that visualization tools can provide valuable insights into the behavior of constraint-based solvers.  ...  variables through the awakening of all other constraints.  ... 
doi:10.1145/1056018.1056022 dblp:conf/softvis/GhoniemCFJ05 fatcat:7lxu5h672za7vjwidvmcfhkkxq

On the Direct Kinematics Problem of Parallel Mechanisms

Arthur Seibel, Stefan Schulz, Josef Schlattmann
2018 Journal of Robotics  
This number of solutions can be further reduced by introducing additional constraints, for example, combinatorial or planarity constraints [9] .  ...  The measurement vector m that includes the raw data of the acceleration sensors, the gyroscopes, and the magnetic sensors is sent to the sensor fusion filter, for example, a Kalman filter.  ... 
doi:10.1155/2018/2412608 fatcat:rpbmxoccgfg2pcnjdo62gnwzuu

AutoCompress: An Automatic DNN Structured Pruning Framework for Ultra-High Compression Rates

Ning Liu, Xiaolong Ma, Zhiyuan Xu, Yanzhi Wang, Jian Tang, Jieping Ye
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Structured weight pruning is a representative model compression technique of DNNs to reduce the storage and computation requirements and accelerate inference.  ...  As an example, AutoCompress outperforms the prior work on automatic model compression by up to 33× in pruning rate (120× reduction in the actual parameter count) under the same accuracy.  ...  As a key property, ADMM can effectively deal with a subset of combinatorial constraints and yield optimal (or at least high quality) solutions.  ... 
doi:10.1609/aaai.v34i04.5924 fatcat:ahz42fbqobaeff74kvvlbgisdi

Non-Structured DNN Weight Pruning – Is It Beneficial in Any Platform? [article]

Xiaolong Ma, Sheng Lin, Shaokai Ye, Zhezhi He, Linfeng Zhang, Geng Yuan, Sia Huat Tan, Zhengang Li, Deliang Fan, Xuehai Qian, Xue Lin, Kaisheng Ma (+1 others)
2020 arXiv   pre-print
We urge the community not to continue the DNN inference acceleration for non-structured sparsity.  ...  Enhancing Solution Feasibility and High Solution Quality In problem (1), the constraint is combinatorial.  ...  Luckily, the associated constraints in the DNN weight pruning and quantization belong to this subset of combinatorial constraints, making ADMM applicable to DNN mode compression.  ... 
arXiv:1907.02124v2 fatcat:axz7kmo7sje2tjn6apqprnqeim

Editorial for the special issue: Knowledge-based systems and data science

Hamido Fujita, Ali Selamat
2018 Applied intelligence (Boston)  
This special issue highlights the presentation of recent theoretical, operational and real-world applications on multiple industries benefitted through the implementation of artificial intelligence techniques  ...  from the extended versions of the 29th International Conference on Industrial, Engineering & Other Applications of Applied Intelligent Systems (IEA/AIE 2016, Morioka, Japan) and the submitted papers through  ...  The paper titled "A multi-level approach to ubiquitous modeling and solving constraints in combinatorial optimization problems in production and distribution" presents a multi-level approach to the modeling  ... 
doi:10.1007/s10489-018-1143-0 fatcat:cmbrbvje75fwze5ohj6m5utghu
« Previous Showing results 1 — 15 out of 5,119 results