Filters








3,816 Hits in 2.4 sec

Effective Parallelisation for Machine Learning [article]

Michael Kamp and Mario Boley and Olana Missura and Thomas Gärtner
2018 arXiv   pre-print
This is a significant step towards a general answer to an open question on the efficient parallelisation of machine learning algorithms in the sense of Nick's Class (NC).  ...  We present a novel parallelisation scheme that simplifies the adaptation of learning algorithms to growing amounts of data as well as growing needs for accurate and confident predictions in critical applications  ...  Introduction This paper contributes a novel and provably effective parallelisation scheme for a broad class of learning algorithms.  ... 
arXiv:1810.03530v1 fatcat:kijomnorwzbmlpe47oxofa7qn4

ParallelPC: an R package for efficient constraint based causal exploration [article]

Thuc Duy Le, Tao Hoang, Jiuyong Li, Lin Liu, Shu Hu
2015 arXiv   pre-print
The package is not only suitable for super-computers or clusters, but also convenient for researchers using personal computers with multi core CPUs.  ...  In this paper, we present an R package, ParallelPC, that includes the parallelised versions of these causal exploration algorithms.  ...  As PC-simple (PC-Select) is efficient in small datasets, we use the Adult dataset from UCI Machine Learning Repository with 48842 samples.  ... 
arXiv:1510.03042v1 fatcat:kjsgqivgojdihbsnet6nlfmt5y

New Machine Learning Developments in ROOT/TMVA

Kim Albertsson, Sergei Gleyzer, Marc Huwiler, Vladimir Ilievski, Lorenzo Moneta, Saurav Shekar, Victor Estrade, Akshay Vashistha, Stefan Wunsch, Omar Andres Zapata Mesa, A. Forti, L. Betev (+3 others)
2019 EPJ Web of Conferences  
The Toolkit for Multivariate Analysis, TMVA, the machine learning package integrated into the ROOT data analysis framework, has recently seen improvements to its deep learning module, parallelisation of  ...  Both dense and convo-lutional layers are shown to be competitive on small-scale networks suitable for high-level physics analyses in both training and in single-event evaluation.  ...  TMVA is a ROOT-integrated framework for machine learning.  ... 
doi:10.1051/epjconf/201921406014 fatcat:zitvj5ym25dm3nobk5y4wdl2ii

Online Non-stationary Boosting [chapter]

Adam Pocock, Paraskevas Yiapanis, Jeremy Singer, Mikel Luján, Gavin Brown
2010 Lecture Notes in Computer Science  
We evaluate the new algorithm against Online Boosting, using the STAGGER dataset and three challenging datasets derived from a learning problem inside a parallelising virtual machine.  ...  Oza's Online Boosting algorithm provides a version of Ad-aBoost which can be trained in an online way for stationary problems.  ...  We are investigating the application of ML techniques to automatic parallelisation problems, running inside a Java virtual machine.  ... 
doi:10.1007/978-3-642-12127-2_21 fatcat:f5mnp7vfuva5dnkifmt6gjnrom

Parallel ILP for distributed-memory architectures

Nuno A. Fonseca, Ashwin Srinivasan, Fernando Silva, Rui Camacho
2008 Machine Learning  
This has brought into focus machine learning techniques like Inductive Logic Programming (ILP) that are able to extract human-comprehensible models for complex relational data.  ...  The growth of machine-generated relational databases, both in the sciences and in industry, is rapidly outpacing our ability to extract useful information from them by manual means.  ...  Acknowledgements The authors would like to acknowledge the anonymous reviewers for the insightful and detailed comments that greatly improved the paper.  ... 
doi:10.1007/s10994-008-5094-2 fatcat:kbow55zlozcfngmzl4rz37l4k4

Characterising Across-Stack Optimisations for Deep Convolutional Neural Networks [article]

Jack Turner, José Cano, Valentin Radu, Elliot J. Crowley, Michael O'Boyle, Amos Storkey
2018 arXiv   pre-print
Since such systems are where some of their most useful applications lie (e.g. obstacle detection for mobile robots, vision-based medical assistive technology), significant bodies of work from both machine  ...  learning and systems communities have attempted to provide optimisations that will make CNNs available to edge devices.  ...  The authors are grateful to Lizhong Chen and the anonymous reviewers for their valuable contributions.  ... 
arXiv:1809.07196v1 fatcat:wxevr5hprveiro5lg2aie5nnem

Characterising Across-Stack Optimisations for Deep Convolutional Neural Networks

Jack Turner, Jose Cano, Valentin Radu, Elliot J. Crowley, Michael OrBoyle, Amos Storkey
2018 2018 IEEE International Symposium on Workload Characterization (IISWC)  
Since such systems are where some of their most useful applications lie (e.g. obstacle detection for mobile robots, vision-based medical assistive technology), significant bodies of work from both machine  ...  learning and systems communities have attempted to provide optimisations that will make CNNs available to edge devices.  ...  The authors are grateful to Lizhong Chen and the anonymous reviewers for their valuable contributions.  ... 
doi:10.1109/iiswc.2018.8573503 dblp:conf/iiswc/TurnerCRCOS18 fatcat:hxxhuovm6fhyhheg55vtwyvsoi

An Enhanced Parallelisation Model for Performance Prediction of Apache Spark on a Multinode Hadoop Cluster

Nasim Ahmed, Andre L. C. Barczak, Mohammad A. Rashid, Teo Susnjak
2021 Big Data and Cognitive Computing  
In this paper, we proposed two distinct parallelisation models for performance prediction.  ...  Our insight is that each node in a Hadoop cluster can communicate with identical nodes, and a certain function of the non-parallelisable runtime can be estimated accordingly.  ...  The major advantage of Apache Spark for machine learning is its end-to-end capabilities.  ... 
doi:10.3390/bdcc5040065 fatcat:pc4q65uwzfdv5lfmlfhftohaeq

The Need for Low Bias Algorithms in Classification Learning from Large Data Sets [chapter]

Damien Brain, Geoffrey I. Webb
2002 Lecture Notes in Computer Science  
Sampling and parallelisation have proved useful means for reducing computation time when learning from large data sets.  ...  This paper reviews the appropriateness for application to large data sets of standard machine learning algorithms, which were mainly developed in the context of small data sets.  ...  Bias and Variance What other fundamental properties of machine learning algorithms are required for learning from large data sets?  ... 
doi:10.1007/3-540-45681-3_6 fatcat:dw6wrhtz3jd27l77w4pudeyb4u

A parallel genetic algorithms framework based on Hadoop MapReduce

Filomena Ferrucci, Pasquale Salza, M-Tahar Kechadi, Federica Sarro
2015 Proceedings of the 30th Annual ACM Symposium on Applied Computing - SAC '15  
This paper describes a framework for developing parallel Genetic Algorithms (GAs) on the Hadoop platform, following the paradigm of MapReduce.  ...  Subject The "Chicago Crime" dataset (from the UCI Machine Learning Repository was used. The dataset has 13 features and 10000 instances.  ...  GAs are usually executed on single machines as sequential programs, so scalability issues prevent that they are effectively applied to real-world problems.  ... 
doi:10.1145/2695664.2696060 dblp:conf/sac/FerrucciSKS15 fatcat:tbr3ib3y6jb2ze7iux3mukuci4

Nonlinear Behaviour in the MPI-Parallelised Model of the Rat Somatosensory Cortex

Grzegorz M. Wojcik, Wieslaw A. Kaminski
2008 Informatica  
Because of a high degree of complexity effective parallelisation of algorithms is required.  ...  We propose method of parallelisation for the network and the results of simulations using GENESIS parallelised for MPI environment are presented. An occurrence of nonlinear behaviour is demonstrated.  ...  Acknowledgements This work has been supported by the Maria Curie-Sklodowska University, Lublin, Poland (under the grant of UMCS Vice President 2007) and Polish State Committee for Scientific Research under  ... 
doi:10.15388/informatica.2008.224 fatcat:oubkm3lyd5gepepqo2mvy5kxoa

Decision tree building on multi-core using FastFlow

Marco Aldinucci, Salvatore Ruggieri, Massimo Torquati
2013 Concurrency and Computation  
TORQUATI Quinlan [3], a cornerstone in data mining and machine learning (see e.g., [4] ).  ...  Nevertheless, the potential for improvements is vast, and it resides in the idle CPU cores on the user's machine.  ...  We thank the Competence Center Gateway for HPC of the IT Center, University of Pisa, for the use of the Magny-Cours box.  ... 
doi:10.1002/cpe.3063 fatcat:cejcokjvmrg5dg4e53e3qpbesq

Scaling up classification rule induction through parallel processing

Frederic Stahl, Max Bramer
2012 Knowledge engineering review (Print)  
Parallelisation seems to be a natural and cost effective way to scale up data mining technologies.  ...  This paper surveys advances in parallelisation in the field of classification rule induction.  ...  of the n machines there is a learning algorithm L installed that learns a local concept out of the data samples locally stored on each machine.  ... 
doi:10.1017/s0269888912000355 fatcat:wbm4gvuu4jbsxoe2qmvl5ae6ne

Page 26 of Journal of Research and Practice in Information Technology Vol. 26, Issue 1 [page]

1994 Journal of Research and Practice in Information Technology  
It also describes an experimental methodology for use in measuring the effects of speed up learning Chapter 3, On Integrating Machine Learning with Planning by Gerald DeJong et al, and Chapter 4, The Role  ...  Although the reports do mention a variety of fronts of machine learning research, the book as a whole does not provide a general picture of the machine learning area.  ... 

Measuring and Evaluating Parallel State-Space Exploration Algorithms

Jonathan Ezekiel, Gerald Lüttgen
2008 Electronical Notes in Theoretical Computer Science  
We discuss and answer these questions based on our experience with parallelising Saturation -a symbolic algorithm for generating state-spaces of asynchronous system models -on a shared-memory architecture  ...  Doing so will hopefully spare newcomers to the growing PDMC community from having to learn these lessons the hard way, as we did over a painful period of almost three years.  ...  Larger shared-memory machines can offer more processors for performance evaluation but are less readily available.  ... 
doi:10.1016/j.entcs.2007.10.020 fatcat:zk5odpq7cnhmpgfqwlyuweqrqy
« Previous Showing results 1 — 15 out of 3,816 results