325,051 Hits in 6.3 sec

A network-based model for high-dimensional information filtering

Nikolaos Nanas, Manolis Vavalis, Anne De Roeck
2010 Proceeding of the 33rd international ACM SIGIR conference on Research and development in information retrieval - SIGIR '10  
The Vector Space Model has been and to a great extent still is the de facto choice for profile representation in contentbased Information Filtering.  ...  Experiments show that the network profile representation can more effectively capture additional information about a user's interests and thus achieve significant performance improvements over a vector-based  ...  Instead, we proposed an alternative network-based model for profile representation that, in the case of textual information, captures correlations between terms appearing in the same context.  ... 
doi:10.1145/1835449.1835485 dblp:conf/sigir/NanasVR10 fatcat:y27kbjl2d5g2beg5xfhqivprve

Graph Representation-Based Deep Multi-View Semantic Similarity Learning Model for Recommendation

Jiagang Song, Jiayu Song, Xinpan Yuan, Xiao He, Xinghui Zhu
2022 Future Internet  
However, traditional social network-based collaborative filtering algorithms will encounter problems such as low recommendation performance and cold start due to high data sparsity and uneven distribution  ...  To this end, this paper proposes a collaborative filtering recommendation algorithm based on graphsage (GraphSAGE-CF).  ...  Graph embedding models embed large-scale information networks into a low-dimensional space, where a low-dimensional feature vector represents each network node.  ... 
doi:10.3390/fi14020032 fatcat:kfd5r7lwyvb7ndlh7dv35hvbdy

Multi-Band Multi-Resolution Fully Convolutional Neural Networks for Singing Voice Separation [article]

Emad M. Grais, Fei Zhao, Mark D. Plumbley
2019 arXiv   pre-print
In this paper, we propose a multi-band multi-resolution fully convolutional neural network (MBR-FCN) for singing voice separation.  ...  Deep neural networks with convolutional layers usually process the entire spectrogram of an audio signal with the same time-frequency resolutions, number of filters, and dimensionality reduction scale.  ...  filters and high dimensionality reduction scale is applied on the extracted features.  ... 
arXiv:1910.09266v1 fatcat:izwmh6qybzc7tk4bfaizevhb5m

Automated Filter Pruning based on High-dimensional Bayesian Optimization

Taehyeon Kim, Heungjun Choi, Yoonsik Choe
2022 IEEE Access  
To address this issue, we propose a high-dimensional Bayesian optimization-based filter pruning (HDBOFP) algorithm, which aims to automatically determine the most appropriate pruning rate for each convolutional  ...  Filter pruning is necessary to efficiently deploy convolutional neural networks on edge devices that have limited computational resources and power budgets.  ...  To overcome this limitation, the proposed algorithm utilizes a low-dimensional embedding-based Bayesian optimization, called high-dimensional Bayesian optimization.  ... 
doi:10.1109/access.2022.3153025 fatcat:3ilvdujvgjbx3jd7n2hl6yhqnm

Data Filter Function Incremental Mining based on Feature Selection in an Active Distribution Network

Song Deng, Qingyuan Cai, Zi Zhang, Lechan Yang, Tinglei Huang, Changan Yuan
2020 IET Cyber-Physical Systems  
And then, based on feature selection, the authors propose a data filtering function model mining algorithm by using gene expression programming (DFFM-FSGEP).  ...  Therefore, feature selection algorithm based on rough set is first given to reduce the complexity of massive and high dimensional data.  ...  Algorithm description To design a better data distributed intelligent filtering model, this paper proposes a fast feature selection algorithm based on rough set for massive, high-dimensional and distributed  ... 
doi:10.1049/iet-cps.2019.0094 fatcat:jeeehyje6nfjxolvzqa3fkhzvm

Combination of two-dimensional cochleogram and spectrogram features for deep learning-based ASR

Andros Tjandra, Sakriani Sakti, Graham Neubig, Tomoki Toda, Mirna Adriani, Satoshi Nakamura
2015 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
This paper explores the use of auditory features based on cochleograms; two dimensional speech features derived from gammatone filters within the convolutional neural network (CNN) framework.  ...  Performance was evaluated in the framework of hybrid neural network -hidden Markov model (NN-HMM) system on TIMIT phoneme sequence recognition task.  ...  Fig. 1 . 1 a) Spectrogram b) Cochleogram Fig. 2 . 2 a) Low level feature combination for DNN b) Low level features combination for CNN Fig. 3 . 3 a) High level feature combination for DNN b) High level  ... 
doi:10.1109/icassp.2015.7178827 dblp:conf/icassp/TjandraSNTAN15 fatcat:rwpcq7xdwnenlhlf6rtoyaseiy

Netboost: Boosting-supported network analysis improves high-dimensional omics prediction in acute myeloid leukemia and Huntington's disease [article]

Pascal Schlosser and Jochen Knaus and Maximilian Schmutz and Konstanze Döhner and Christoph Plass and Lars Bullinger and Rainer Claus and Harald Binder and Michael Lübbert and Martin Schumacher
2019 arXiv   pre-print
First, a boosting-based filter is combined with the topological overlap measure to identify the essential edges of the network.  ...  It is a freely available Bioconductor R package for dimension reduction and hypothesis generation in high-dimensional omics applications.  ...  The results published here are in part based upon data generated by the TCGA Research Network: https://www.  ... 
arXiv:1909.12551v1 fatcat:vv3zjbvrlbdnzngqxe5fb43nve

Smart Modeling of Microwave Devices

Humayun Kabir, Lei Zhang, Ming Yu, Peter Aaen, John Wood, Qi-Jun Zhang
2010 IEEE Microwave Magazine  
This makes neural networks a useful choice for device modeling where a mathematical model is not available. The evaluation from input to output of a neural network model is also very fast.  ...  Neural networks learn device data through an automated training process, and the trained neural networks are then used as fast and accurate models for efficient high-level circuit and system design.  ...  We describe the high dimensional neural network modeling through an example of Coupling matrix Synthesis Filter Specification Neural Inverse Model for Filter Ideal Coupling Values Geometrical Dimensions  ... 
doi:10.1109/mmm.2010.936079 fatcat:qg6zdzfpjzcuxkqswitorjxahq

Task and Spatial Frequency Effects on Face Specialization

Matthew N. Dailey, Garrison W. Cottrell
1997 Neural Information Processing Systems  
frequency network shows a strong specialization for faces.  ...  When one module receives low spatial frequency information and the other receives high spatial frequency information, and the task is to identify the faces while simply classifying the objects, the low  ...  We showed that a model based on the mixture of experts architecture, in which a gating network implements competitive learning between two simple homogeneous modules, could develop a specialization such  ... 
dblp:conf/nips/DaileyC97 fatcat:klo2ihjzybbvxojhvzddhvjubq

A comparison of deep learning and linear-nonlinear cascade approaches to neural encoding [article]

Theodore H. Moskovitz, Nicholas A. Roy, Jonathan W. Pillow
2018 bioRxiv   pre-print
We show that models with nonlinearities parametrized by deep networks achieve higher accuracy for a fixed number of filters, and can extract a larger number of informative filters than traditional models  ...  Finally, we perform a dimensionality analysis of LNP models trained with deep learning methods, revealing that a large number of filters are needed to accurately describe the neural responses of many cells  ...  estimators, even for simple white noise stimuli; (iv) We show that early visual neural responses are in fact high-dimensional, in contrast to previous assumptions, and that DNN-based LNP models are better  ... 
doi:10.1101/463422 fatcat:u654b5fb2vhhlclfuw5dgdbgsa

Implementation of Short Video Click-Through Rate Estimation Model Based on Cross-Media Collaborative Filtering Neural Network

Ying Feng, Guisheng Zhao, Gengxin Sun
2022 Computational Intelligence and Neuroscience  
In this paper, we analyze the construction of cross-media collaborative filtering neural network model to design an in-depth model for fast video click-through rate projection based on cross-media collaborative  ...  filtering neural network.  ...  filtering neural network model based on Deep FM for fast video click-through rate estimation.  ... 
doi:10.1155/2022/4951912 pmid:35685157 pmcid:PMC9173947 fatcat:tq3tdoj5qzha3ltxvpaz7thge4

Analog Circuit Fault Diagnosis Using a Novel Variant of aConvolutional Neural Network

Liang Han, Feng Liu, Kaifeng Chen
2021 Algorithms  
Aiming to accurately diagnose the faults of analog circuits, this paper proposes a novel variant of a convolutional neural network, namely, a multi-scale convolutional neural network with a selective kernel  ...  In MSCNN-SK, a multi-scale average difference layer is developed to compute multi-scale average difference sequences, and then these sequences are taken as the input of the model, which enables it to mine  ...  Two-dimensional scatter distribution of original data and features extracted from network model in Four-opamp biquad high-pass filter circuit formed by T-SNE method. (a) fault data.  ... 
doi:10.3390/a15010017 fatcat:wpft7xzaefctlcd2gpnonpet6a

A Deep 2D Convolutional Network for Waveform-Based Speech Recognition

Dino Oglic, Zoran Cvetkovic, Peter Bell, Steve Renals
2020 Interspeech 2020  
To mitigate that and allow for learning of robust models, we propose a deep 2D convolutional network in the waveform domain.  ...  The first layer of the network decomposes waveforms into frequency sub-bands, thereby representing them in a structured high-dimensional space.  ...  In addition to having a more flexible inductive bias such a model would be less susceptible to the information loss that is inherent to waveform compression by means of a projection to a lower dimensional  ... 
doi:10.21437/interspeech.2020-1870 dblp:conf/interspeech/OglicC0R20 fatcat:p6o45vsy5jejjoooodls3dmylm

Two-Stage Latent Dynamics Modeling and Filtering for Characterizing Individual Walking and Running Patterns with Smartphone Sensors

Jaein Kim, Juwon Lee, Woongjin Jang, Seri Lee, Hongjoong Kim, Jooyoung Park
2019 Sensors  
For the task of characterizing movements, the proposed method makes use of encoding the high-dimensional sequential data from movements into random variables in a low-dimensional latent space.  ...  In this paper, we propose a machine learning approach, referred to as 'two-stage latent dynamics modeling and filtering' (TS-LDMF) method, where we combine a latent space modeling stage with a nonlinear  ...  In the backbone structure, we use the transition network for a process model in low-dimensional latent space, and the emitter network for a measurement model for sensors (e.g., [12, 24] ).  ... 
doi:10.3390/s19122712 fatcat:y6gik36rnrhhtlzht52xmks2l4

Research on Vision Measurement System of Mechanical Workpiece Based on Machine Vision

Yubin Guo
2022 Highlights in Science, Engineering and Technology  
With the rapid development of modern industrial industry, the workpiece measurement system of many industrial products gradually exceeds the operating load, and the operating efficiency is not high, the  ...  In order to solve the problem of automatic and intelligent measurement of mechanical workpieces in industrial production, an automatic mechanical workpiece measurement system based on machine vision is  ...  The mathematical model of Gaussian filtering for images in a two-dimensional coordinate system is as follows.  ... 
doi:10.54097/hset.v1i.428 fatcat:i3v24nlxf5bfva3p6tz2igw2oa
« Previous Showing results 1 — 15 out of 325,051 results