Filters








205 Hits in 7.6 sec

Optimizing Neural Network Classifiers with ROOT on a Rocks Linux Cluster [chapter]

Tomas Lindén, Francisco García, Aatos Heikkinen, Sami Lehti
Applied Parallel Computing. State of the Art in Scientific Computing  
We present a study to optimize multi-layer perceptron (MLP) classification power with a Rocks Linux cluster [1] .  ...  We report on the performance of PROOF and on the integration of PROOF with the cluster environment in use and on the physics performance of new neural classifiers developed in this study.  ...  Conclusions ROOT and PROOF have been installed on two Linux clusters and integrated with the NPACI Rocks Linux Cluster Distribution and the Sun N1 Grid Engine batch queue system.  ... 
doi:10.1007/978-3-540-75755-9_124 dblp:conf/para/LindenGHL06 fatcat:sn42skkrsfcynfklejh4u3tyxe

Testing TMVA software in b-tagging for the search of MSSM Higgs bosons at the LHC

T Lampén, F Garcia, A Heikkinen, P Kaitaniemi, V Karimäki, M J Kortelainen, S Lehti, T Lindén, L Wendl
2008 Journal of Physics, Conference Series  
Achievable b tagging efficiency is studied with more than ten MVA classifiers at 1% mistagging rate. Most classifiers were found to perform better than the simple track counting algorithm.  ...  We test the usage of a Toolkit for Multivariate Data Analysis (TMVA) in b tagging.  ...  Acknowledgments A. Heikkinen  ... 
doi:10.1088/1742-6596/119/3/032028 fatcat:x56qtd242rg45bznchiwuc5otq

From Imitation to Prediction, Data Compression vs Recurrent Neural Networks for Natural Language Processing

Juan Andres Laura, Gabriel Omar Masi, Luis Argerich
2018 Inteligencia Artificial  
In our journey, a fundamental difference between a Data Compression Algorithm and Recurrent Neural Networks has been discovered.  ...  If this is possible, then the problem comes down to determining if a compression algorithm is even more intelligent than a neural network in such tasks.  ...  Alex Graves for his comments on an earlier attempt to make handwriting recognition with PAQ compressor.  ... 
doi:10.4114/intartif.vol21iss61pp30-46 fatcat:rwqclrrxnfconh6yvq3jhl4k64

Commodity single board computer clusters and their applications

Steven J. Johnston, Philip J. Basford, Colin S. Perkins, Herry Herry, Fung Po Tso, Dimitrios Pezaros, Robert D. Mullins, Eiko Yoneki, Simon J. Cox, Jeremy Singer
2018 Future generations computer systems  
For instance, certain types of neural networks [110, 111] are being adapted to run effectively on SBC clusters, with low power constraints, limited memory usage, and minimal GPU processing.  ...  Alpine has a read-only root partition with dynamic overlays, which mitigates problems with SD card corruption.  ...  He works on programming languages and runtime systems, with a particular focus on many core platforms. He has published more than 30 papers in these areas. Dr.  ... 
doi:10.1016/j.future.2018.06.048 fatcat:u55aojod3bgqtkxnfppw2zhtje

Hershel

Zain Shamsi, Ankur Nandwani, Derek Leonard, Dmitri Loguinov
2014 The 2014 ACM international conference on Measurement and modeling of computer systems - SIGMETRICS '14  
Since these techniques have not been studied before, we first pioneer stochastic theory of single-packet OS fingerprinting, build a database of 116 OSs, design a classifier based on our models, evaluate  ...  This can be overcome by approaches that rely on a single SYN packet to elicit a vector of features from the remote server.  ...  A vast majority of the hosts are clustered on the values just before the initial TTL defaults 64, 128, and 255.  ... 
doi:10.1145/2591971.2591972 dblp:conf/sigmetrics/ShamsiNLL14 fatcat:qznjviajjve67ckzypqfid6yly

Hershel

Zain Shamsi, Ankur Nandwani, Derek Leonard, Dmitri Loguinov
2014 Performance Evaluation Review  
Since these techniques have not been studied before, we first pioneer stochastic theory of single-packet OS fingerprinting, build a database of 116 OSs, design a classifier based on our models, evaluate  ...  This can be overcome by approaches that rely on a single SYN packet to elicit a vector of features from the remote server.  ...  A vast majority of the hosts are clustered on the values just before the initial TTL defaults 64, 128, and 255.  ... 
doi:10.1145/2637364.2591972 fatcat:iciignl3g5bhvl2xapzgb7b3cu

A roadmap of clustering algorithms: finding a match for a biomedical application

B. Andreopoulos, A. An, X. Wang, M. Schroeder
2008 Briefings in Bioinformatics  
We compare algorithms on the basis of desirable clustering features, and outline algorithms' benefits and drawbacks as a basis for matching them to biomedical applications.  ...  Clustering is ubiquitously applied in bioinformatics with hierarchical clustering and k-means partitioning being the most popular methods.  ...  Another disadvantage of model-based clustering (especially neural networks) is slow processing time on large datasets.  ... 
doi:10.1093/bib/bbn058 pmid:19240124 fatcat:5vesuqp7cjhhxhsrlzn4vm4djq

Systematic Literature Review over IDPS, Classification and Application in its Different Areas

Shehroz Afzal, Jamil Asim
2021 STATISTICS, COMPUTING AND INTERDISCIPLINARY RESEARCH  
In this Survey paper focuses on Classifying various kinds of IDS with the major types of attacks based on intrusion methods.  ...  Rock solid network security is a major challenge that can be overcome by strengthening the network against threats such as hackers, malware, botnets, data thieves, etc.  ...  Genetic algorithms are a heuristic approach to optimization, based on the principles of evolution.  ... 
doi:10.52700/scir.v3i2.58 fatcat:xrczlxjg5ncclf2ftxyw3y5zce

Efficient disease detection in gastrointestinal videos – global features versus neural networks

Konstantin Pogorelov, Michael Riegler, Sigrun Losada Eskeland, Thomas de Lange, Dag Johansen, Carsten Griwodz, Peter Thelin Schmidt, Pål Halvorsen
2017 Multimedia tools and applications  
Furthermore, it is built in a modular way, so that it can be easily extended to deal with other types of abnormalities.  ...  The system combines deep learning neural networks, information retrieval, and analysis of global and local image features in order to implement multi-class classification, detection and localization.  ...  creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a  ... 
doi:10.1007/s11042-017-4989-y fatcat:w5adpg2k6jc3hoc6rai5mcxwj4

Algorithmic Clustering of Music [article]

Rudi Cilibrasi , Ronald de Wolf
2003 arXiv   pre-print
We present a fully automatic method for music classification, based only on compression of strings that represent the music pieces.  ...  It is based on an ideal theory of the information content in individual objects (Kolmogorov complexity), information distance, and a universal similarity metric.  ...  ] , Bayesian classifiers [6] , hidden Markov models [5] , ensembles of nearest-neighbor classifiers [7] or neural networks [6, 13] .  ... 
arXiv:cs/0303025v1 fatcat:bljpvxu7dvdz7glcreap6qotiy

LIRE

Mathias Lux, Michael Riegler, Pål Halvorsen, Konstantin Pogorelov, Nektarios Anagnostopoulos
2016 Proceedings of the 7th International Conference on Multimedia Systems - MMSys '16  
We have researched and developed a holistic medical multimedia system addressing a use case with an important medical and societal impact.  ...  Then, we designed and developed a set of lesion and findings detection and localization approaches based on hand-crafted methods as well as on global-, local-and deepfeature-based methods, which serves  ...  creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a  ... 
doi:10.1145/2910017.2910630 dblp:conf/mmsys/LuxRHPA16 fatcat:jfweprlxfzb2lommgkw6nqqwi4

Artificial Intelligence [article]

Denis Rothman
2018 Zenodo  
In May 2017, Google revealed AutoML, an automated machine learning system that could create an artificial intelligence solution without the assistance of a human engineer.  ...  We have to find a way to classify those dots with a neural network. Step 1 Defining a feedforward neural network We look at our piece of paper. We don't have a computer.  ...  : The ability of a neural network to make non-separable information separable and classifiable represents one of the core powers of deep learning.  ... 
doi:10.5281/zenodo.5599649 fatcat:vyqr3zzw5rfzpp3zifezw7uzxq

Clustering by compression [article]

Rudi Cilibrasi and Paul Vitanyi (CWI and University of Amsterdam)
2004 arXiv   pre-print
We present a new method for clustering based on compression.  ...  A theoretical precursor, the normalized information distance, co-developed by one of the authors, is provably optimal but uses the non-computable notion of Kolmogorov complexity.  ...  the clustering process.  ... 
arXiv:cs/0312044v2 fatcat:7kxq7xadyzd2xnup36aqxrnboq

TRUST-TECH based Methods for Optimization and Learning [article]

Chandan K. Reddy
2007 arXiv   pre-print
Many problems that arise in machine learning domain deal with nonlinearity and quite often demand users to obtain global optimal solutions rather than local optimal ones.  ...  on several test systems.  ...  Training Neural Networks Without loss of generality, we consider a feedforward neural network with one input layer, one hidden layer and one output layer.  ... 
arXiv:0712.4126v1 fatcat:ibcu3fjjzngzlf25ow5c4hfvkm

Clustering by Compression

R. Cilibrasi, P.M.B. Vitanyi
2005 IEEE Transactions on Information Theory  
We present a new method for clustering based on compression.  ...  A theoretical precursor, the normalized information distance, co-developed by one of the authors, is provably optimal.  ...  visualization of the clustering process.  ... 
doi:10.1109/tit.2005.844059 fatcat:imrem6kkszbfjhh6aom6udfhga
« Previous Showing results 1 — 15 out of 205 results