Filters








6,112 Hits in 3.7 sec

Going Deep: Graph Convolutional Ladder-Shape Networks

Ruiqi Hu, Shirui Pan, Guodong Long, Qinghua Lu, Liming Zhu, Jing Jiang
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
To solve this problem, we present graph convolutional ladder-shape networks (GCLN), a novel graph neural network architecture that transmits messages from shallow layers to deeper layers to overcome the  ...  While they have achieved great success in semi-supervised node classification on graphs, current approaches suffer from the over-smoothing problem when the depth of the neural networks increases, which  ...  Graph Convolutional Ladder-shape Networks Figure 3 : Illustration of a graph convolutional ladder-shape networks (GCLNs).  ... 
doi:10.1609/aaai.v34i03.5673 fatcat:iqkayosfjbeijhy4zkyolhwyxi

Graph Regularized Variational Ladder Networks for Semi-Supervised Learning

Cong Hu, Xiao-Ning Song
2020 IEEE Access  
TABLE 1 : 1 Description of Convolutional Graph Regularized Variational Ladder Networks GRVLN (for MNIST) GRVLN (for CIFAR and SVHN) Input: 28×28×1 Output: 32×32×3 5×5 conv. 32, ReLU 3×3 conv. 96  ...  It can be seen from the experimental results that the convolutional Laplace ladder network still has better classification ability than the convolutional ladder network and other semi-supervised deep learning  ...  He was a Visiting Researcher with the Centre for Vision, Speech, and Signal Processing, University of Surrey, Guildford, U.K., from 2014 to 2015.  ... 
doi:10.1109/access.2020.3038276 fatcat:onzpdkdlyfc2bflpqkma4gedc4

Graph Representation Learning via Ladder Gamma Variational Autoencoders

Arindam Sarkar, Nikhil Mehta, Piyush Rai
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
We present a probabilistic framework for community discovery and link prediction for graph-structured data, based on a novel, gamma ladder variational autoencoder (VAE) architecture.  ...  Our framework is also fairly modular and can leverage a wide variety of graph neural networks as the VAE encoder.  ...  attention based architecture for node classification on graph-structured data.  ... 
doi:10.1609/aaai.v34i04.6013 fatcat:be2y5orhzbhntd5oeckulc5wtm

Relational Graph Neural Network Design via Progressive Neural Architecture Search [article]

Ailing Zeng, Minhao Liu, Zhiwei Liu, Ruiyuan Gao, Jing Qin, Qiang Xu
2022 arXiv   pre-print
We propose a novel solution to addressing a long-standing dilemma in the representation learning of graph neural networks (GNNs): how to effectively capture and represent useful information embedded in  ...  We verify the proposed LADDER-GNN on seven diverse semi-supervised node classification datasets. Experimental results show that our solution achieves better performance than most existing GNNs.  ...  To verify the effectiveness of the proposed GNN representation learning solution, we demonstrate it on seven semisupervised node classification datasets for both homogeneous and heterogeneous graphs with  ... 
arXiv:2105.14490v4 fatcat:7qzj7xuwpnbubfpwempt6a3m3q

Application of a K-Ladder Connectivity Algorithm for Clustering of Protein Evolutionary Network

Reshma Nibhani, Avi Soffer, Ahuva Mu'alem, Zeev Volkovich, Zakharia Frenkel
2014 International Journal of Modeling and Optimization  
This graph can be used for detecting hidden relatedness between proteins, which is highly significant in protein annotation. Effective EN analysis requires an appropriate graph clustering approach.  ...  Index Terms-K-ladder, connectivity algorithm, network clustering, protein evolutionary network, formatted protein sequence space, protein-protein interaction networks. .  ...  ACKNOWLEDGMENT The authors are grateful to Miki Dabush and Svetlana Kolodiy for help in preliminary calculations.  ... 
doi:10.7763/ijmo.2014.v4.403 fatcat:qy4nwyhrizfxlpcvdjvyolhcku

Recurrent Ladder Networks [article]

Isabeau Prémont-Schwarz, Alexander Ilin, Tele Hotloo Hao, Antti Rasmus, Rinu Boney, Harri Valpola
2017 arXiv   pre-print
We propose a recurrent extension of the Ladder networks whose structure is motivated by the inference required in hierarchical latent variable models.  ...  We present results for fully supervised, semi-supervised, and unsupervised tasks.  ...  Acknowledgments We would like to thank Klaus Greff and our colleagues from The Curious AI Company for their contribution in the presented work, especially Vikram Kamath and Matti Herranen.  ... 
arXiv:1707.09219v4 fatcat:whv3oifnivgapnnf34cnheevrq

Keep Your Bearings: Lightly-Supervised Information Extraction with Ladder Networks That Avoids Semantic Drift

Ajay Nagesh, Mihai Surdeanu
2018 Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers)  
We propose a novel approach to semisupervised learning for information extraction that uses ladder networks (Rasmus et al., 2015) .  ...  We empirically demonstrate the superior performance of our system compared to the state-of-the-art on two standard datasets for named entity classification.  ...  Conclusion We discussed a novel application of ladder networks to the task of lightly supervised named entity classification.  ... 
doi:10.18653/v1/n18-2057 dblp:conf/naacl/NageshS18 fatcat:jaugabpgjralvngqiys6h7jzsi

Cost-Sensitive and Sparse Ladder Network for Software Defect Prediction

Jing SUN, Yi-mu JI, Shangdong LIU, Fei WU
2020 IEICE transactions on information and systems  
Taking the above issues into account, we propose a cost-sensitive and sparse ladder network (CSLN) for SDP.  ...  We firstly introduce the semi-supervised ladder network to extract the deep feature representations.  ...  As we know, SDP is a binary classification problem, ladder network may be applied in SDP.  ... 
doi:10.1587/transinf.2019edl8198 fatcat:fw2omth7h5af5crk5zasm22yye

EvoNet: A Neural Network for Predicting the Evolution of Dynamic Graphs [article]

Changmin Wu, Giannis Nikolentzos, Michalis Vazirgiannis
2020 arXiv   pre-print
Neural networks for structured data like graphs have been studied extensively in recent years. To date, the bulk of research activity has focused mainly on static graphs.  ...  Specifically, we use a graph neural network along with a recurrent architecture to capture the temporal evolution patterns of dynamic graphs.  ...  For ladder graph, we observe a mismatch at the beginning of the sequence for small size graphs but then a coincidence of the two lines on large size graphs.  ... 
arXiv:2003.00842v1 fatcat:izakvtzharbufg6sztjlm7nsim

Identification of b-/y-ions in MS/MS spectra using a two stage neural network

James P Cleveland, John R Rose
2013 Proteome Science  
Our approach uses a staged neural network to model ion fragmentation patterns and estimate the posterior probability of each ion type.  ...  If we select too few peaks then the ion-ladder interpretation of the spectrum will contain gaps that can only be explained by permutations of combinations of amino acids.  ...  Hans Wessels for his assistance with the MS/MS dataset. Declarations The publication costs for this article were funded by the corresponding author.  ... 
doi:10.1186/1477-5956-11-s1-s4 pmid:24565419 pmcid:PMC3907776 fatcat:qwqkzotrprd3zdavbkhssk7eqm

Challenge-Aware Traffic Protection in Wireless Mobile Backhaul Networks [chapter]

Javier Martín-Hernández, Christian Doerr, Johannes Lessmann, Marcus Schöller
2012 Lecture Notes in Computer Science  
Challenges in real networks can have very different characteristics how they impact the network, e.g., unusual high load can temporarily overload the network in a particular area, whereas a heavy rain  ...  thunderstorm cell affects all wireless links while moving across the network area, and last attacks exploiting software bugs affect systems spread all over the network.  ...  During network operation the Graph Explorer assesses possible placements of rope-ladders such as to be maximally robust towards certain challenges.  ... 
doi:10.1007/978-3-642-30039-4_5 fatcat:e2e5fyvxize3vh3eihyx6m6pli

A review of various semi-supervised learning models with a deep learning and memory approach

Jamshid Bagherzadeh, Hasan Asil
2018 Iran Journal of Computer Science  
A research solution for future studies is to benefit from memory to increase such an effect. Memory-based neural networks are new models of neural networks which can be used in this area.  ...  Therefore, semi-supervised learning is more practical and useful for solving most of the problems.  ...  Different models were presented for this method of supervised data such as deep generative models, virtual adversarial, and ladder.  ... 
doi:10.1007/s42044-018-00027-6 fatcat:nccifurxyzc33fa5xfprrlupxq

Semi-supervised long short-term memory for human action recognition

Hong Liu, Chang Liu, Runwei Ding
2020 The Journal of Engineering  
One is the LSTM ladder network, the other is the Symmetrical LSTM network.  ...  For example, the minimal cut of graphs [20] of Gaussian random field [21] , graph theory, graph-based semi-supervised learning and so on.  ...  Inspired by the ladder network, we design the LSTM ladder network, which consists of three sub-networks: a noisy encoder, a clean encoder and a decoder as shown in Fig. 4.  ... 
doi:10.1049/joe.2019.1166 fatcat:6v3nodzd6beyngjge5saosgjma

AUTOMATIC CONSTRUCTION OF HIERARCHICAL ROAD NETWORKS

Weiping Yang
2016 ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
The enrichment on road network data is important to a high successful rate of feature matching for road networks and to geospatial data integration.  ...  The method is based on a pattern graph which maintains nodes and paths as junctions and through-traffic roads.  ...  For instance, to target a ladder pattern, there most exist a series of similar straight lines, ladder steps, connected to a common line, a ladder bar, with T-intersects.  ... 
doi:10.5194/isprs-annals-iii-2-37-2016 fatcat:zl27lr5425c6logshnoii6bile

Counting representable sets on simple graphs

Peter G. Jeavons
1993 Discrete Applied Mathematics  
This paper derives exact expressions for the number of representable sets when the corresponding graph is cycle-free or series-parallel.  ...  A set of colourings which is the solution to some network of specified constraints is said to be a representable set.  ...  Hence, the values available for each individual constraint are more tightly restricted on these ladder graphs than on any of the graphs previously considered.  ... 
doi:10.1016/0166-218x(93)90150-m fatcat:axoyyht4vngyfmumxe6h65ic24
« Previous Showing results 1 — 15 out of 6,112 results