IJCNN 2014

180 papers
The 2014 International Joint Conference on Neural Networks (IJCNN) featured 227 papers covering a broad range of topics in computational intelligence and machine learning. Prominent research directions for this year included computer vision techniques for distinguishing real objects from mirrored reflections and interactive perception for robotic material and state discovery. Prolific authors contributing to the proceedings included Junbin Gao, Licheng Jiao, Badong Chen, and Dongbin Zhao. The technical program highlighted advancements in classification methods, such as dual instance and attribute weighting for Naive Bayes and ensemble methods for multi-label data streams in non-stationary environments. Other research focused on hardware implementations, including the KLMS algorithm using FPGA, and robust data recovery through Tensor Low-Rank Representation. Additional studies addressed word-level multi-script identification from video frames and fundamental algorithms for finding convex hull vertices in metric space. This edition reflected a diverse focus on both theoretical neural network architectures and practical applications in robotics, signal processing, and pattern recognition.

Top Research Directions

Research directions most represented by IJCNN 2014 papers.

Supervised Dimensionality Reduction 1
Discriminating Between Non-Orthogonal Quantum States 1
Learning Kinematics and Intuitive Physics Through Motor Babbling 1
Disambiguating Physical Space from Specular Reflections 1
Input Adaptive Compute Allocation for Language Models 1
Manifold Regularization for Semi-Supervised Learning 1
Overcoming Euclidean Distortion with Hyperbolic Embeddings 1
Latent Factor Discovery and Split Sample Validation 1
Understanding and Mitigating Decoherence in Open Quantum Systems 1
Self-Explainable Neural Networks without Performance Loss 1
Guiding Semi Supervised Learning with Consistency Constraints 1
Exploiting the Latent Structure of Unlabeled Data 1
The Experimental Validation Gap in Advanced Control Systems 1

Top Authors

AuthorPapers
Junbin Gao5
Licheng Jiao4
Aurelio Uncini3
Badong Chen3
Dongbin Zhao3
Furao Shen3
Michael Blumenstein3
Shuyuan Yang3
Trung Le3
Alessandro Sperduti2
Arindam Basu2
Artur Garcez2
Bao-liang Lu2
Bo Yuan2
Chang Liu2
Chengqi Zhang2
Dacheng Tao2
Danilo Comminiello2
Dinh Phung2
Gang Pan2

Impactful Papers

Papers from IJCNN 2014 that shaped research clusters.

Foundational

Influential

Papers

Title
Single channel single trial P300 detection using extreme learning machine: Compared with BPNN and SVM
Single image super-resolution via learned representative features and sparse manifold embedding
Smartphone battery saving by bit-based hypothesis spaces and local Rademacher Complexities
Soft-constrained nonnegative matrix factorization via normalization
Spatial filter adaptation based on geodesic-distance for motor EEG classification
Speaker verification with deep features
Spectral clustering-based local and global structure preservation for feature selection
Spike-timing dependent morphological learning for a neuron with nonlinear active dendrites
Stability of Hopfield neural networks with event-triggered feedbacks
STDP learning rule based on memristor with STDP property
Structured sparse coding method for infrared small target detection in video sequence
Supervised Bayesian sparse coding for classification
Supervised topic regression via experts
SVD truncation schemes for fixed-size kernel models
Tensor LRR based subspace clustering
The HC calculus, quaternion derivatives and caylay-hamilton form of quaternion adaptive filters and learning systems
The Parzen kernel approach to learning in non-stationary environment
The state of the art of memristive neural systems: Models and applications
Trimmed affine projection algorithms
Unmanned aerial vehicles (UAV) heading optimal tracking control using online kernel-based HDP algorithm
Unsupervised robust Bayesian feature selection
User daily activity pattern learning: A multi-memory modeling approach
User-generated-video summarization using Sparse Modelling
Using Chou's amphiphilic Pseudo-Amino Acid Composition and Extreme Learning Machine for prediction of Protein-protein interactions
Using EEG artifacts for BCI applications
Using HDDT to avoid instances propagation in unbalanced and evolving data streams
Using recurrent networks for non-temporal classification tasks
Using self-organizing incremental neural network (SOINN) For radial basis function networks
Vessel segmentation in retinal images with a multiple kernel learning based method
Visual saliency via loss coding

« prev Page 4/4