A Proposal of Neuron Filter: A Constraint Resolution Scheme of Neural Networks for Combinatorial Optimization Problems
スポンサーリンク
概要
- 論文の詳細を見る
A constraint resolution scheme in the Hopfield-type neural network named "Neuron Filter" is presented for efficiently solving combinatorial optimization problems. The neuron filter produces an output that satisfies the constraints of the problem as best as possible according to both neuron inputs and outputs. This paper defines the neuron filter and shows its introduction into existing neural networks for N-queens problems and FPGA board-level routing problems. The performance is evaluated through simulations where the results show that our neuron filter improves the searching capability of the neural network with the shorter computation time.
- 社団法人電子情報通信学会の論文
- 2000-09-25
著者
-
Higashino Teruo
Department Of Informatics And Mathematical Science Osaka University
-
Higashino Teruo
The Department Of Informatics And Mathematical Science Graduate School Of Engineering Science Osaka
-
Funabiki N
Graduate School Of Natural Science And Technology Okayama University
-
Takenaka Yoichi
The Department Of Informatics And Mathematical Science Graduate School Of Engineering Science Osaka
-
TAKENAKA Yoichi
Division of Informatics and Mathematical Science, Graduate School of Engineering Science, Osaka Univ
-
FUNABIKI Nobuo
Division of Informatics and Mathematical Science, Graduate School of Engineering Science, Osaka Univ
-
HIGASHINO Teruo
Division of Informatics and Mathematical Science, Graduate School of Engineering Science, Osaka Univ
関連論文
- A WDS Clustering Algorithm for Wireless Mesh Networks
- An Optical-Drop Wavelength Assignment Algorithm for Efficient Wavelength Reuse under Heterogeneous Traffic in WDM Ring Networks(Discrete Mathematics and Its Applications)
- A Minimum Dead Space Algorithm for Generalized Isochronous Channel Reuse Problems in DQDB Networks(Network)
- P2PMM_router : A Two-Stage Heuristic Algorithm to Peer-to-Peer Multicast Routing Problems in Multihome Networks(Discrete Mathematics and Its Applications)
- A Quasi-Solution State Evolution Algorithm for Channel Assignment Problems in Cellular Networks(Special Section on Discrete Mathematics and Its Applications)
- Time-Action Alternating Model for Timed Processes and Its Symbolic Verification of Bisimulation
- A Proposal of Optimal Path Selection Algorithm for Static and Mobile Multicast Routing Problems
- Deriving Concurrent Synchronous EFSMs from Protocol Specifications in LOTOS (Special Section on Selected Papers from the 11th Workshop on Circuits and Systems in Karuizawa)
- A Method to Convert Concurrent EFSMs with Multi-Rendezvous into Synchronous Sequential Circuit(Special Section on Concurrent Systems Technology)
- Execution Time Analysis for Binary Code Executed on a Pipelined Processor Using Parametric Model Checking
- Relaxation of Coefficient Sensitiveness to Performance for Neural Networks Using Neuron Filter through Total Coloring Problems
- A Proposal of Neuron Filter: A Constraint Resolution Scheme of Neural Networks for Combinatorial Optimization Problems
- A Digital Neural Network for Multilayer Channel Routing with Crosstalk Minimization
- A Massive Digital Neural Network for Total Coloring Problems
- A Gradual Neural Network Approach for Time Slot Assignment in TDM Multicast Switching Systems
- Comparisons of Energy-Descent Optimization Algorithms for Maximum Clique Problems (Special Section on Discrete Mathematics and Its Applications)
- A Minimal-State Processing Search Algorithm for Graph Coloring Problems