Neural Information Processing

Name: Neural Information Processing | 25th International Conference, ICONIP 2018, Siem Reap, Cambodia, December 13-16, 2018, Proceedings, Part II
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

25th International Conference, ICONIP 2018, Siem Reap, Cambodia, December 13-16, 2018, Proceedings, Part II

Long Cheng Andrew Chi Sing Leung Seiichi Ozawa(Editor)

Springer (Publisher)

Published on 3. December 2018

XXII, 735 pages

E-Book

PDF with digital watermarking

System requirements

978-3-030-04179-3 (ISBN)

€53.49incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Intro
Preface
ICONIP 2018 Organization
Contents - Part II
Other Neural Network Models
Improved Kernel Density Estimation Self-organizing Incremental Neural Network to Perform Big Data Analysis
Abstract
1 Introduction
2 Related Works
2.1 Kernel Density Estimation
2.2 Self-organizing Incremental Neural Network
2.3 KDESOINN
3 Proposed Method
4 Experimental Study
4.1 Fixed Gaussian Distribution
4.2 Changing Gaussian Distribution
5 Conclusion
References
HISBmodel: A Rumor Diffusion Model Based on Human Individual and Social Behaviors in Online Social Networks
1 Introduction
2 Related Work and Preliminary Knowledge
2.1 Rumor Propagation Problem in OSNs
2.2 Damped Harmonic Motion
3 Proposed Rumor Propagation Model
3.1 Individual Behavior Toward a Rumor Formulation
3.2 Individual Opinion Formulation
3.3 Human Social Interactions Rules
3.4 Rumor Propagation Process
4 Rumor Influence Minimization Strategy Formulation
5 Experiments
5.1 Model Performance
5.2 Performance of the Proposed Strategy
6 Conclusions
References
Multi-scale Feature Decode and Fuse Model with CRF Layer for Boundary Detection
1 Introduction
2 Related Work
3 MSDF Model
3.1 Network Architecture
3.2 Learning and Deployment
4 CRF Model Based on MSDF
4.1 CRF Model Architecture
5 Experiments
6 Conclusion
References
Sentimental Analysis for AIML-Based E-Health Conversational Agents
1 Introduction
2 Harlie an E-Health Chat-Bot
3 Methodology
3.1 Graph-Based Sentiment Model
3.2 Machine Learning-Based Sentiment Model
3.3 Data Corpus
4 Results
5 Discussion
6 Conclusions
References
Hashtag Recommendation with Attention-Based Neural Image Hashtagging Network
1 Introduction
2 Related Works
2.1 Hashtag & Tag Recommendation
2.2 Attention Based Hashtag Recommendation
3 The Proposed Method
3.1 Encoder with Sequential Attention
3.2 Decoder for Hashtag Recommendation
3.3 Training
3.4 Hashtag Recommendation
4 Experiments
4.1 Experiments Settings
4.2 Baseline Model
4.3 Comparison Experiments
4.4 Model Component Impact Analysis
5 Conclusions
References
CocoNet: A Deep Neural Network for Mapping Pixel Coordinates to Color Values
1 Introduction
2 Related Work
3 Method
3.1 Intuition
3.2 CocoNet
4 Experiments
4.1 Data Sets
4.2 Evaluation
4.3 Image Denoising Results
4.4 Image Resampling Results
4.5 Image Completion Results
5 Conclusion and Further Work
References
BCMLP: Binary-Connected Multilayer Perceptrons
1 Introduction
2 Related Work
2.1 Sparse Connection
2.2 Sensitivity Evaluation
3 BCMLP
4 Sensitivity Analysis
5 Experiments
5.1 Amount of Parameters
5.2 Sensitivity Test
5.3 Noisy Logic Gates
6 Conclusion
References
Network of Recurrent Neural Networks: Design for Emergence
1 Introduction
2 Background
2.1 Systems Theory and Emergence
2.2 Systems Theory and Recurrent Neural Networks
3 Network of Recurrent Neural Networks
3.1 Overall Architecture
3.2 Methodology i: Aggregation
3.3 Methodology ii: Specialization
3.4 Information-Theoretical Measurements
4 Task-Based Evaluations
4.1 Handwriting Classification
4.2 Named Entity Recognition
4.3 Question Classification
5 Information-Theoretical Evaluations
6 Conclusion
References
Computationally Efficient Radial Basis Function
1 Introduction
2 RBF Networks and RBF Kernels
3 Square Non-linear Radial Basis Function (SQ-RBF)
4 Experiments and Analysis
4.1 SinE Function Approximation
4.2 Nonlinear Dynamic System Identification
4.3 Mackey-Glass Time Series Prediction
4.4 Inverse Cosine Function Approximation
5 Conclusion
References
A Model for Age and Gender Profiling of Social Media Accounts Based on Post Contents
1 Introduction
2 Related Works
3 Methodology
3.1 Data Collection
3.2 Feature Retrieval
3.3 Model Production
3.4 Evaluation
4 Results and Discussion
5 Conclusion
References
A Wiener Causality Defined by Relative Entropy
1 Introduction
2 Notations
2.1 Residual Errors
2.2 KL Divergence
3 REC: Relative Entropy Causality
4 Experiments
4.1 Stationary MVAR
4.2 Non-stationary MVAR
4.3 Time-Varing Nonstationary MVAR
4.4 Predictive Models
5 Discussions and Conclusions
References
Scene Graph Generation Based on Node-Relation Context Module
Abstract
1 Introduction
2 Related Work
2.1 Baseline Scene Graph Generation
2.2 Attention Models
2.3 Relationship Referring
3 Node-Relation Context Module
3.1 Attention Module for Relationship Prediction
3.2 Fusion Function
4 Experiments
4.1 Evaluation Metrics
4.2 Ablation Study
4.3 Comparison with Existing Works
4.4 Qualitative Analysis
5 Conclusions
Acknowledgement
References
Cross-Layer Convolutional Siamese Network for Visual Tracking
1 Introduction
2 Related Work
3 Method
3.1 Cross-Layer Convolutional Response Maps
3.2 Adaptive contrastive loss
4 Implementation Details
5 Experiments
6 Conclusion
References
A Data Augmentation Model Based on Variational Approach
1 Introduction
2 Motivation
3 Related Work
3.1 Factored Gated Restricted Boltzmann Machine
3.2 Variational Autoencoder
4 Variational Factored Gated Restricted Boltzmann Machine
4.1 Model Description
4.2 Inference
5 Experiment
5.1 MNIST
5.2 CIFAR-10
6 Conclusion
References
Asynchronous Value Iteration Network
1 Introduction
2 Background
2.1 Markov Decision Process
2.2 Learning and Planning
2.3 Asynchronous Methods
3 Asynchronous Value Iteration Network
3.1 VIN
3.2 Asynchronous VI Module
4 Experiments
5 Conclusion and Future Work
References
FVR-SGD: A New Flexible Variance-Reduction Method for SGD on Large-Scale Datasets
1 Introduction
2 Background
2.1 Variance Reduction
2.2 Distributed SGD
3 FVR-SGD Algorithm
4 Convergence Analysis
5 Distributed Scheme
6 Experiments and Evaluations
6.1 Single Machine
6.2 Distributed Mode
7 Conclusion
References
A Neural Network Model for Gating Task-Relevant Information by Rhythmic Oscillations
Abstract
1 Introduction
2 Model
2.1 Visual Tasks
2.2 Overview of Our Model
2.3 The Model of V1
2.4 The Model of V2
2.5 Synaptic Weights of V1 and V2 Neurons
3 Results
3.1 Tuning Modulations of V1 Neurons by Top-Down Influence
3.2 A Role of Rhythmic Oscillations in Gating Task-Relevant Information
4 Conclusion
References
A Hybrid Model Based on the Rating Bias and Textual Bias for Recommender Systems
1 Introduction
2 Preliminary
3 Proposed Model
4 Experiments
4.1 Datasets
4.2 Experimental Setup
4.3 Evaluation Metrics
4.4 Experimental Results
5 Conclusion
References
Phase and Amplitude Modulation in a Neural Oscillatory Model of the Orientation Map
1 Introduction
2 Methods
2.1 The Network Architecture
2.2 Oscillatory Neural Response Demodulation
2.3 Phase Demodulation
2.4 Amplitude Demodulation
3 Results and Discussion
4 Conclusion
References
Neural Networks Models for Analyzing Magic: The Gathering Cards
1 Introduction
1.1 Color
1.2 Card Type
2 Our Goals
3 Methods
3.1 Environment
4 Classifying MTG Card Illustrations by Color
5 Classifying MTG Card Illustrations by Type
6 Text Classification
7 Matching Randomly Generated Card Text to Input Images
7.1 Card Generation Examples
8 Discussion and Future Work
References
MulAttenRec: A Multi-level Attention-Based Model for Recommendation
1 Introduction
2 Related Work
2.1 Attention-Based Mechanism in RS
2.2 Combination of FM and Deep Neural Networks
3 Our Approach
3.1 Overview
3.2 Word-Level Attention-Based Mechanism
3.3 Full-Text Level Attention-Based Mechanism
3.4 Hybrid Prediction Layer
3.5 Learning
4 Experiment
4.1 Experimental Settings
4.2 Hyper-Parameter Study
4.3 Performance Evaluation
4.4 Effect of Multi-level Attention
4.5 Effect of Hybrid Prediction Layer
4.6 Keyword and Review Analysis
5 Conclusion and Future Work
References
SSteGAN: Self-learning Steganography Based on Generative Adversarial Networks
1 Introduction
2 Related Work
2.1 Generative Adversarial Networks
2.2 Steganography
3 Self-learning Steganography Based on Generative Adversarial Networks
3.1 Model Design and Objective Function
3.2 Structure Details
4 Experiment
4.1 Data Preparation and Parameters Setting
4.2 Experimental Results
4.3 Comparison with Related Work
4.4 Decryption Security Evaluation
5 Conclusion
References
A Multidimensional Interaction-Focused Model for Ad-Hoc Retrieval
1 Introduction
2 The Model
3 Experiment Methodology
3.1 Dataset
3.2 Baseline Methods
3.3 Evaluation Method
4 Experiments Details
4.1 Training Details
4.2 Experiment Results
5 Analysis of MIF
5.1 Matching Tensor Dimension
5.2 Error Analysis
6 Related Work
7 Conclusion
References
Accounting Results Modelling with Neural Networks: The Case of an International Oil and Gas Company
Abstract
1 Introduction
2 Related Works
3 Neural Network Modelling of Accounting Results
3.1 Accounting Results of the Case Company
3.2 Selection of Accounting Results as NN Outputs
3.3 Selection of Business Factors as NN Inputs
3.4 Data Collection and Pre-processing
3.5 NN Architectures and Training
4 Results
5 Conclusion
References
Attention Based Dialogue Context Selection Model
1 Introduction
2 Model
2.1 Framework
3 Experiments
3.1 Evaluation Metrics and Results
4 Conclusion
References
Memory-Based Model with Multiple Attentions for Multi-turn Response Selection
1 Introduction
2 Related Work
3 The Approach
3.1 Model Overview
3.2 Matching Vectors Layer
3.3 Memory Building Layer
3.4 Multiple Attentions Layer
3.5 Matching Prediction
4 Experiments
4.1 Datasets
4.2 Model Training
4.3 Experimental Results
4.4 Case Study
5 Conclusion
References
A Robust LPNN Technique for Target Localization Under Hybrid TOA/AOA Measurements
1 Introduction
2 General Background
2.1 Time of Arrival Model
2.2 Angle of Arrival (AOA) Model
2.3 Lagrange Programming Neural Network (LPNN)
3 The Proposed Approach
3.1 Problem Formulation
3.2 Approximation of l1 Norm
3.3 Network Stability Study
4 Simulation
4.1 Setting
4.2 Case 1
4.3 Case 2
5 Conclusion
References
Central Pattern Generator Based on Interstitial Cell Models Made from Bursting Neuron Models
Abstract
1 Introduction
2 Interstitial Cell Model
3 CPG Based on Interstitial Cell Models
4 Conclusion
References
Active Feedback Framework with Scan-Path Clustering for Deep Affective Models
1 Introduction
2 Experiment Settings
2.1 Experiment with Attention
2.2 Attention-Deprived Experiment
3 Method
3.1 Attention Evaluation
3.2 EEG Data Filtering for Emotion Recognition
4 Experimental Results
4.1 Results of Attention Evaluation
4.2 Results of Feedback
5 Conclusion
References
Stability Analysis
A New Robust Stability Result for Delayed Neural Networks
1 Introduction
2 Preliminaries
3 Existence and Uniqueness Analysis
4 Stability Analysis
5 Conclusions
References
A Novel Criterion for Global Asymptotic Stability of Neutral-Type Neural Networks with Discrete Time Delays
1 Introduction
2 Global Stability Analysis
3 Conclusions
References
Anti-synchronization of Neural Networks with Mixed Delays
1 Introduction
2 Preliminaries
3 Main Results
4 Numerical Example
5 Conclusions
References
Incremental Stability of Neural Networks with Switched Parameters and Time Delays via Contraction Theory of Multiple Norms
1 Introduction
2 Preliminaries
3 Main Results
4 Numerical Example
5 Conclusions
References
Lag Synchronization of Complex-Valued Neural Networks with Time Delays
1 Introduction
2 Preliminaries
3 Main Results
4 Numerical Example
5 Conclusion
References
Continuous Attractors of Nonlinear Neural Networks with Asymmetric Connection Weights
1 Introduction
2 Preliminaries
3 Representations of Continuous Attractors
4 Simulations
5 Conclusions
References
Optimization
Two Matrix-Type Projection Neural Networks for Solving Matrix-Valued Optimization Problems
1 Introduction
2 Matrix-Valued Optimization Problem
3 Matrix-Type Projection Neural Network
4 Stability Results
5 Computed Results
6 Conclusion
References
An Adaptive Ant Colony System for Public Bicycle Scheduling Problem
Abstract
1 Introduction
2 Description of PBSP
2.1 Distance Model
2.2 Time Window Model
3 AACS to Solve PBSP
3.1 Key Components in Designing AACS
3.2 Complete AACS Algorithm
4 Experiments and Discussions
4.1 Comparison Without Bicycle Constraints
4.2 Experimental Results in Distance Model
4.3 Experimental Results in Time Window Model
5 Conclusion
Acknowledgments
References
An Artificial Neural Network for Distributed Constrained Optimization
1 Introduction
2 Problem Statement and Optimization Algorithm
3 Consensus Analysis
4 Numerical Example
5 Conclusion
References
An Estimation of Distribution Algorithm for Large-Scale Optimization with Cooperative Co-evolution and Local Search
1 Introduction
2 Relate Work
2.1 Estimation of Distribution Algorithm
2.2 Cooperative Co-evolution
2.3 Differential Grouping Strategy
3 Estimation of Distribution Algorithm with CC Framework
3.1 Distribution Estimation
3.2 Cooperative Co-evolution
3.3 Local Search
4 Experiment Studies
4.1 Parameter Setting
4.2 Experiment Result
5 Conclusion
References
A Collaborative Neurodynamic Approach to Symmetric Nonnegative Matrix Factorization
1 Introduction
2 Preliminaries
2.1 Problem Formulation
2.2 Particle Swarm Optimization
2.3 Collaborative Neurodynamic Optimization
3 Main Results
3.1 Augmented Lagrangian Method
3.2 Model Analysis
4 Experimental Results
5 Conclusions
References
Modularity Maximization for Community Detection Using Genetic Algorithm
Abstract
1 Introduction
2 Our Proposed Method
2.1 Problem Definition
2.2 Random Walk Distance
2.3 Community Structure Representation
2.4 Genetic Algorithm
2.5 Comparison of Proposed Method and Ga-Net
3 Experimental Results
4 Conclusion
Acknowledgments
References
Continuous Trade-off Optimization Between Fast and Accurate Deep Face Detectors
1 Introduction
2 Related Work
3 Methodology
4 Experiments
4.1 Data Sets
4.2 Evaluation Details
4.3 Results and Discussion
5 Conclusion
References
Multi-Dimensional Optical Flow Embedded Genetic Programming for Anomaly Detection in Crowded Scenes
1 Introduction
2 Related Work
2.1 Genetic Programming
2.2 Histogram of Oriented Optical Flow
3 Our Approach for Anomaly Detection in Crowed Scenes
3.1 MDOF and Feature Extraction
3.2 MDOF Embedded GP Evolution
4 Experiments
4.1 Dataset
4.2 Results on the UMN Data Set
4.3 Results on the UCSD Data Set
5 Conclusions
References
Robust Regression with Nonconvex Schatten p-Norm Minimization
1 Introduction
2 Related Work
2.1 Robust Regression
2.2 Schatten p-Norm and p-Norm
3 Robust Regression Combined with Joint Schatten p-Norm and p-Norm
3.1 The Proposed Model
3.2 Optimization
3.3 Convergence Analysis
4 Experiment
4.1 Face Recognition on YaleB Dataset
4.2 Face Recognition on PIE Dataset
4.3 Object Recognition on COIL20 Dataset
4.4 Performance of Varying p
5 Conclusion
References
Adaptive Crossover Memetic Differential Harmony Search for Optimizing Document Clustering
Abstract
1 Introduction
2 Related Work
3 Document Clustering Using ACMDHS Optimization
4 Initial Parameters Setting and Test Results
4.1 Parameter Settings
4.2 Test Results
4.2.1 The Internal and External Evaluation
5 Conclusions
Acknowledgements
References
Neurodynamics-Based Nonnegative Matrix Factorization for Classification
Abstract
1 Introduction
2 Related Works
2.1 Non-negative Matrix Factorization
2.2 Gaussian Process Regression (GPR)
2.3 Support Vector Machine (SVM)
2.4 Enhanced K-Nearest Neighbor (ENN)
3 Background
3.1 Continuous-Time Projection Neural Network
3.2 Discrete-Time Projection Neural Network
4 Neurodynamics-Based Non-negative Matrix Factorization Method
4.1 Neurodynamics Equation of Discrete-Time Projection Neural Network (DTPNN)
4.2 Backtracking Line Search
4.3 Neurodynamics-Based Non-negative Matrix Factorization Algorithm
4.4 Combined NMF and Classification Algorithm
5 Experimental Results
5.1 Initialization
5.2 Convergent Objective Function Values
5.3 CPU Running Time
5.4 Convergent Objective Function Values vs. Iterations
5.5 Classification Results
6 Conclusions
Acknowledgements
References
A Multi-kernel Semi-supervised Metric Learning Using Multi-objective Optimization Approach
1 Introduction
2 Background
2.1 Kernel Learning with Relative Distance
3 Problem Statement
4 Multi-kernel Based Semi-supervised Metric Learning Using Multi-objective Optimization
4.1 Chromosome Representation
4.2 Genetic Operators
4.3 Objective Functions
4.4 NSGA-III
4.5 Selection of a Single Solution for Reporting
5 Experimental Results
5.1 Data Sets Used
5.2 Evaluation Measures
5.3 Discussion of Results
6 Conclusion
References
Software-as-a-Service Composition in Cloud Computing Using Genetic Algorithm
1 Introduction
2 Related Work
3 The SaaS Composition Problem
4 Our Genetic Algorithm
4.1 Genetic Encoding
4.2 Genetic Operators
4.3 Fitness Function
5 Evaluation
5.1 Performance and Scalability with Number of Tasks
5.2 Performance and Scalability with Number of Selectable SaaS Components per Task
5.3 Summary of the Experimental Results
6 Conclusion
References
Evolving Computationally Efficient Hashing for Similarity Search
1 Introduction
2 Overview
2.1 Nearest Neighbor Search and Hashing
2.2 Random Key Genetic Algorithms
3 Materials and Methods
3.1 Distance Preserving Sparse Projections
3.2 Biased Variable Length Random Key Genetic Algorithm
3.3 Interpreter Alleles and Encoding
3.4 Biased Variable Length Crossover
3.5 BVLRKGA
4 Experiments
4.1 Genome Length Adaptation
4.2 Learning to Hash
5 Conclusions
References
Combining Two-Phase Local Search with Multi-objective Ant Colony Optimization
Abstract
1 Introduction
2 Background
2.1 Two-Objective Traveling Salesman Problem (bTSP)
2.2 Iterated Local Search - Variable Neighborhood Search (ILS-VNS)
2.3 Two-Phase Local Search (TPLS)
3 Proposed Algorithm
3.1 Related Works
3.2 Hybrid MOACO + TPLS
3.3 TPLS + WLS / PLS
3.4 ILS-VNS
4 Conclusion
References
A Neural Network Based Global Optimal Algorithm for Unconstrained Binary Quadratic Programming Problem
1 Introduction
2 The Upper Bound Algorithm and Its Recurrent Neural Network Model
3 The Lower Bound Algorithm
4 Neural Network Based Algorithm to Sort xk
5 The Algorithm and Experimental Results
6 Conclusion
References
Supervised Learning
Stochasticity-Assisted Training in Artificial Neural Network
1 Introduction
2 Concepts
2.1 Absolute Error Based Noise
2.2 Rate of Change of Error
3 Implementation Details
4 Experimental Results
4.1 Protocol 1 Noise Decision Update
4.2 Protocol 2 Noise Decision Update
5 Conclusion
References
An Independent Approach to Training Classifiers on Physiological Data: An Example Using Smiles
Abstract
1 Introduction
2 Smiles and Observers' Physiology
2.1 Smilers
2.2 Observers
2.3 Data Recording
2.4 Signal Processing
2.5 Signal Extraction
3 An Independent Approach
4 Results and Discussion
5 Conclusion
References
Mixed Precision Weight Networks: Training Neural Networks with Varied Precision Weights
1 Introduction
2 Related Works
2.1 BinaryConnect
2.2 Ternary Weight Networks
3 Mixed Precision Weight Networks
4 Experimental Results
4.1 MNIST
4.2 CIFAR-10
4.3 CIFAR-100
5 Conclusion and Future Works
References
Policy Space Noise in Deep Deterministic Policy Gradient
Abstract
1 Introduction
2 Background
2.1 Markov Decision Process and Reinforcement Learning
2.2 Deep Deterministic Policy Gradient
3 Exploratory Noise in DDPG
3.1 DDPG with Action Space Noise
3.2 DDPG with Policy Space Noise
4 Experiment
4.1 Experimental Settings
4.2 Main Evaluation
4.3 Model Test
5 Conclusion
Acknowledgement
References
Mixup of Feature Maps in a Hidden Layer for Training of Convolutional Neural Network
1 Introduction
2 Related Works
2.1 Deep Convolutional Neural Network
2.2 Mixup
2.3 Siamese Network and Triplet Network
3 Mixup of Feature Maps in a Hidden Layer
4 Experiment
4.1 Dataset and Network Architectures
4.2 Comparison of the Classification Accuracy
4.3 Comparison of the Feature Maps in the Hidden Layer
5 Conclusion and Future Works
References
Deep Deterministic Policy Gradient with Clustered Prioritized Sampling
Abstract
1 Introduction
2 Model and Algorithm
2.1 Model Structure
2.2 Algorithm
3 Experiments
3.1 Experiments Details
3.2 Experiments Results
4 Conclusion
Acknowledgements
References
Induced Exploration on Policy Gradients by Increasing Actor Entropy Using Advantage Target Regions
1 Introduction
2 Preliminaries
3 Proposed Algorithm
3.1 Policy Computations and Actor Training Mechanisms
3.2 Computation of Modified Advantage Targets
3.3 Analysis
4 Experiments
5 Conclusion
References
MCP Based Noise Resistant Algorithm for Training RBF Networks and Selecting Centers
1 Introduction
2 Background
2.1 RBF Networks Under Multiplicative Noise
2.2 ADMM
3 Development of ADMM-MCP
3.1 ADMM-MCP Algorithm
3.2 Convergence Analysis
4 Simulations
4.1 Settings and Properties of ADMM-MCP Algorithm
4.2 Performance Comparison
5 Conclusion
References
Fault-Resistant Algorithms for Single Layer Neural Networks
1 Introduction
2 ELM and Weight Noise Model
3 Weight Deviation Tolerant Learning
3.1 WDT-IELM Algorithm
3.2 WDTC-IELM Algorithm
4 Simulation
4.1 Setting
4.2 Test Set Error Versus the Number of Hidden Nodes
4.3 Comparison
5 Conclusion
References
Adversarial Minimax Training for Robustness Against Adversarial Examples
1 Introduction
2 Adversarial Attack
2.1 Generation Algorithms of Adversarial Examples
2.2 Adversarial Attack Scenarios
3 Adversarial Minimax Training
4 Computer Simulation
4.1 Setup
4.2 Results
5 Conclusions
References
SATB-Nets: Training Deep Neural Networks with Segmented Asymmetric Ternary and Binary Weights
1 Introduction
1.1 Binary Weight Networks and Model Compression
1.2 Ternary Weight Networks and Model Compression
2 Segmented Asymmetric Ternary and Binary Weights Networks
2.1 Segmentation
2.2 Asymmetric Binary Weights for Fully-Connected(FC) Layers
2.3 Asymmetric Ternary Weights for Convolutional (CONV) Layers
2.4 Heterogeneous Quantized Weights Structure
2.5 Train the SATB-Nets with Stochastic Gradient Descent (SGD) Method
3 Experiments
3.1 VGGNets on CIFAR-10
3.2 AlexNet on ImageNet
4 Conclusion
References
Asynchronous Methods for Multi-agent Deep Deterministic Policy Gradient
Abstract
1 Introduction
2 Related Work
2.1 Reinforcement Learning
2.2 Multi-agent Deep Deterministic Policy Gradient
3 Asynchronous Advantage Multi-agent Deep Deterministic Policy Gradient
4 Experiments
5 Conclusions and Future Work
Acknowledgments
References
A Revisit of Reducing Hidden Nodes in a Radial Basis Function Neural Network with Histogram
Abstract
1 Introduction
2 The Methods
2.1 The RBFNDDA
2.2 The Histogram
2.3 The Hybrid of RBFNDDA and HIST with Radius as the Pruning Indicator
3 Experiment
3.1 Overlapping Computation
3.2 Benchmarked Data Sets and Experimental Setup
3.3 Results and Discussions
4 Conclusion
References
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Neural Information Processing

Description

More details

Other editions

Additional editions

Content

System requirements