Neural Information Processing : 23rd International Conference, ICONIP 2016, Kyoto, Japan, October 16–21, 2016, Proceedings, Part III / / edited by Akira Hirose, Seiichi Ozawa, Kenji Doya, Kazushi Ikeda, Minho Lee, Derong Liu

• Pubblicazione: Cham : , : Springer International Publishing : , : Imprint : Springer, , 2016
• Edizione: 1st ed. 2016.
• Descrizione fisica: 1 online resource (XVIII, 651 p. 215 illus.)
• Disciplina: 006.3
• Soggetto topico: Pattern recognition systems; Computer vision; Artificial intelligence; Computer science; Data mining; Application software; Automated Pattern Recognition; Computer Vision; Artificial Intelligence; Theory of Computation; Data Mining and Knowledge Discovery; Computer and Information Systems Applications
• Persona (resp. second.): HiroseAkira; OzawaSeiichi; DoyaKenji; IkedaKazushi; LeeMinho; LiuDerong
• Nota di contenuto: Intro -- Preface -- Organization -- Contents - Part III -- Time Series Analysis -- Chaotic Feature Selection and Reconstruction in Time Series Prediction -- 1 Introduction -- 2 Chaotic Feature Selection and Reconstruction -- 2.1 Cooperative Neuro-Evolution -- 3 Experiments and Results -- 3.1 Problem Description -- 3.2 Experimental Design -- 3.3 Results and Discussion -- 4 Conclusions and Future Work -- References -- L1/2 Norm Regularized Echo State Network for Chaotic Time Series Prediction -- Abstract -- 1 Introduction -- 2 Echo State Networks -- 3 L1/2 Regularized Echo State Network -- 4 Simulations -- 5 Conclusions -- Acknowledgement -- References -- SVD and Text Mining Integrated Approach to Measure Effects of Disasters on Japanese Economics -- Abstract -- 1 Introduction -- 2 SVD and Text Mining Integrated Approach -- 3 Time Series Stock Data Analysis -- 4 Topic Extraction Results -- 5 Conclusion -- Acknowledgement -- References -- Deep Belief Network Using Reinforcement Learning and Its Applications to Time Series Forecasting -- Abstract -- 1 Introduction -- 2 DBN with BP Learning (The Conventional Method) -- 2.1 DBN with RBM and MLP -- 3 DBN with SGA (The Proposed Method) -- 3.1 The Structure of ANNs with SGA -- 4 Prediction Experiments and Results -- 4.1 CATS Benchmark Time Series Data -- 4.2 Optimization of Meta Parameters -- 4.3 Experiments Result -- 5 Conclusion -- Acknowledgment -- References -- Neuron-Network Level Problem Decomposition Method for Cooperative Coevolution of Recurrent Networks for Time Series Prediction -- 1 Introduction -- 2 Neuron-Network Level Problem Decomposition -- 3 Experimental Setup -- 4 Results -- 5 Discussion -- 6 Conclusion -- References -- Data-Driven Approach for Extracting Latent Features from Multi-dimensional Data -- Yet Another Schatten Norm for Tensor Recovery -- 1 Introduction. ; 2 Theoretical Results -- 2.1 New Norm -- 2.2 Properties -- 2.3 Tensor Recovery -- 3 Experimental Results -- 4 Conclusion -- References -- Memory of Reading Literature in a Hippocampal Network Model Based on Theta Phase Coding -- 1 Introduction -- 2 Computer Simulation -- 3 Results -- 4 Discussion -- References -- Combining Deep Learning and Preference Learning for Object Tracking -- 1 Introduction -- 2 Deep and Preference Learning Tracker -- 2.1 Deep Learning -- 2.2 Preference Learning -- 2.3 Model Update -- 3 Experiments -- 4 Conclusions -- References -- A Cost-Sensitive Learning Strategy for Feature Extraction from Imbalanced Data -- 1 Introduction -- 2 A Motivating Example -- 3 Theoretical Analysis -- 3.1 Imbalance Cost Ratio -- 3.2 Cost-Sensitive Principal Component Analysis (CSPCA) -- 3.3 Cost-Sensitive Non-negative Matrix Factorization (CSNMF) -- 3.4 Revisiting the Motivating Example -- 4 Experiments and Analysis -- 4.1 Experimental Framework -- 4.2 Analysis and Results -- 5 Conclusions and Future Work -- References -- Nonnegative Tensor Train Decompositions for Multi-domain Feature Extraction and Clustering -- 1 Introduction -- 2 Nonnegative Tensor Decomposition Models -- 2.1 NTD Model -- 2.2 NTT Model -- 2.3 NTT-Tucker Model: A Hybrid of the NTD and NTT Models -- 3 NTT-HALS: Proposed Algorithm for NTT and NTT-Tucker -- 3.1 NTT-HALS for NTT -- 3.2 NTT-HALS for NTT-Tucker -- 4 Experiments -- 4.1 Multi-domain Feature Extraction from ERP Data -- 4.2 Feature Extraction and Clustering from ORL Database of Face Images -- 5 Discussion -- References -- Hyper-parameter Optimization of Sticky HDP-HMM Through an Enhanced Particle Swarm Optimization -- 1 Introduction -- 2 Related Work -- 3 Problem Statement -- 4 Method: Hyperparameter Optimization -- 4.1 Random Search -- 4.2 Particle Swarm Optimization -- 4.3 Ring-Based Particle Swarm Optimization. ; 5 Experiment -- 5.1 Dataset -- 5.2 Synthetic Data Results -- 5.3 Tum Kitchen Dataset Results -- 6 Conclusion -- References -- Approximate Inference Method for Dynamic Interactions in Larger Neural Populations -- 1 Introduction -- 2 Methods -- 2.1 The State-Space Model of Neural Interactions -- 2.2 Approximation Methods for a Large-Scale Analysis -- 3 Results -- 4 Conclusion -- References -- Features Learning and Transformation Based on Deep Autoencoders -- 1 Introduction -- 2 Unsupervised Transformation of the Feature Space -- 2.1 Matrix Decomposition and Normalization -- 2.2 Diffusion Maps -- 2.3 Deep Autoencoders -- 3 Topological Clustering -- 4 Experimental Results -- 5 Conclusion -- References -- t-Distributed Stochastic Neighbor Embedding with Inhomogeneous Degrees of Freedom -- 1 Introduction -- 2 Stochastic Neighbor Embedding -- 3 t-Distributed Stochastic Neighbor Embedding -- 4 Inhomogeneous t-SNE -- 4.1 Degrees of Freedom -- 4.2 Cost Function and Its Gradient -- 4.3 Optimization -- 5 Experiments -- 5.1 Experiment 1 -- 5.2 Experiment 2 -- 6 Summary and Discussion -- References -- Topological and Graph Based Clustering Methods -- Parcellating Whole Brain for Individuals by Simple Linear Iterative Clustering -- Abstract -- 1 Introduction -- 2 Materials and Methods -- 2.1 Subjects -- 2.2 Simple Linear Iterative Clustering (SLIC) -- 2.3 Evaluation Metrics -- 3 Experimental Results -- 4 Conclusion and Future Directions -- Acknowledgements -- References -- Overlapping Community Structure Detection of Brain Functional Network Using Non-negative Matrix Factorization -- 1 Introduction -- 2 Methods and Material -- 2.1 Association Matrix Construction with NASR -- 2.2 Community Detection with SNMF -- 2.3 Data Preparation -- 2.4 Experiment -- 3 Results and Discussion -- 3.1 Simulated Data Set -- 3.2 Real Resting-State fMRI Data Set. ; 4 Conclusion and Limitation -- References -- Collaborative-Based Multi-scale Clustering in Very High Resolution Satellite Images -- 1 Introduction -- 2 Multi-scale Communication Between Different Algorithms -- 3 Experimental Results -- 3.1 Description of the Data -- 3.2 Results -- 4 Conclusion -- References -- Towards Ontology Reasoning for Topological Cluster Labeling -- 1 Introduction and Motivations -- 2 Related Work -- 3 Preliminaries About Ontology and Reasoning -- 4 Hybrid Approach: SOM Ontology Based Labeling -- 4.1 Topological Unsupervised Learning Step -- 4.2 Ontology Based Map Labeling -- 5 Experiments -- 5.1 Satellite Images Classification -- 6 Conclusion and Future Work -- References -- Overlapping Community Detection Using Core Label Propagation and Belonging Function -- 1 Introduction -- 2 Label Propagation Algorithm -- 2.1 Standard Label Propagation -- 2.2 Label Propagation with Dams and Core Detection -- 3 Proposed Methods for Detection of Overlapping Communities -- 3.1 Function 1: Membership Function Based on the Density -- 3.2 Function 2 Membership Function Based on the Local Clustering Coefficient -- 3.3 Proposed Community Detection Algorithms -- 4 Evaluation Measures of Community Detection Algorithm, Benchmarks, Experiments and Discussion -- 4.1 Experiments -- 4.2 Comparative Analysis -- 5 Perspectives and Conclusion -- References -- A New Clustering Algorithm for Dynamic Data -- 1 Introduction -- 2 Growing Neural Gas -- 3 A New Two-Level Clustering Algorithm for GNG -- 4 Experimental Results -- 5 Conclusions and Perspectives -- References -- Reinforcement Learning -- Decentralized Stabilization for Nonlinear Systems with Unknown Mismatched Interconnections -- 1 Introduction -- 2 Problem Statement -- 3 Decentralized Controller Design and Stability Analysis -- 3.1 Optimal Control -- 3.2 Neural Network Implementation. ; 3.3 Stability Analysis -- 4 Simulation Study -- 5 Conclusion -- References -- Optimal Constrained Neuro-Dynamic Programming Based Self-learning Battery Management in Microgrids -- 1 Introduction -- 2 Problem Formulation -- 3 Iterative ADP Algorithm for Battery Management System -- 3.1 Derivations of the Iterative ADP Algorithm -- 4 Simulation Analysis -- 5 Conclusion -- References -- Risk Sensitive Reinforcement Learning Scheme Is Suitable for Learning on a Budget -- 1 Introduction -- 2 Incremental Learning on a Budget -- 2.1 Kernel Regression Based Learning on a Budget -- 2.2 Kernel Replacement Algorithm -- 3 Risk Sensitive Reinforcement Learning -- 4 Experimental Results -- 4.1 Results -- 5 Conclusion -- References -- A Kernel-Based Sarsa() Algorithm with Clustering-Based Sample Sparsification -- 1 Introduction -- 2 RL and Kernel Method -- 3 Clustering-Based Selective Kernel Sarsa() -- 3.1 Clustering-Based Novelty Criterion -- 3.2 Framework of CSKS() -- 4 Experiment and Results -- 4.1 Settings of Acrobot -- 4.2 Results -- 5 Conclusion -- References -- Sparse Kernel-Based Least Squares Temporal Difference with Prioritized Sweeping -- 1 Introduction -- 2 Background -- 3 Sparse Kernel-Based Least Squares Temporal Difference with Prioritized Sweeping (PS-SKLSTD) -- 3.1 Sparse Kernel-Based Least Squares Temporal Difference -- 3.2 Kernel-Based Prioritized Sweeping -- 3.3 PS-SKLSTD Algorithm -- 4 Experimental Results -- 4.1 Puddle World -- 4.2 Cart-Pole -- 5 Conclusions -- References -- Computational Intelligence -- Vietnamese POS Tagging for Social Media Text -- 1 Introduction -- 2 Related Work -- 3 Dataset -- 4 Tagging Method -- 5 Experiments -- 5.1 Experimental Setting -- 5.2 Models to Compare -- 5.3 Results -- 6 Conclusion -- References -- Scaled Conjugate Gradient Learning for Quaternion-Valued Neural Networks -- 1 Introduction -- 2 The HR Calculus. ; 3 Conjugate Gradient Algorithms.
• Sommario/riassunto: The four volume set LNCS 9947, LNCS 9948, LNCS 9949, and LNCS 9950 constitues the proceedings of the 23rd International Conference on Neural Information Processing, ICONIP 2016, held in Kyoto, Japan, in October 2016. The 296 full papers presented were carefully reviewed and selected from 431 submissions. The 4 volumes are organized in topical sections on deep and reinforcement learning; big data analysis; neural data analysis; robotics and control; bio-inspired/energy efficient information processing; whole brain architecture; neurodynamics; bioinformatics; biomedical engineering; data mining and cybersecurity workshop; machine learning; neuromorphic hardware; sensory perception; pattern recognition; social networks; brain-machine interface; computer vision; time series analysis; data-driven approach for extracting latent features; topological and graph based clustering methods; computational intelligence; data mining; deep neural networks; computational and cognitive neurosciences; theory and algorithms.
• Titolo autorizzato: Neural Information Processing
• ISBN: 3-319-46675-5
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 9910483725203321
• Lo trovi qui: Univ. Federico II
• Serie: Theoretical Computer Science and General Issues, . 2512-2029 ; ; 9949