Book contents
- Frontmatter
- Contents
- Contributors
- Preface
- 1 Scaling Up Machine Learning: Introduction
- Part One Frameworks for Scaling Up Machine Learning
- 2 MapReduce and Its Application to Massively Parallel Learning of Decision Tree Ensembles
- 3 Large-Scale Machine Learning Using DryadLINQ
- 4 IBM Parallel Machine Learning Toolbox
- 5 Uniformly Fine-Grained Data-Parallel Computing for Machine Learning Algorithms
- Part Two Supervised and Unsupervised Learning Algorithms
- Part Three Alternative Learning Settings
- Part Four Applications
- Subject Index
- References
4 - IBM Parallel Machine Learning Toolbox
from Part One - Frameworks for Scaling Up Machine Learning
Published online by Cambridge University Press: 05 February 2012
Book contents
- Frontmatter
- Contents
- Contributors
- Preface
- 1 Scaling Up Machine Learning: Introduction
- Part One Frameworks for Scaling Up Machine Learning
- 2 MapReduce and Its Application to Massively Parallel Learning of Decision Tree Ensembles
- 3 Large-Scale Machine Learning Using DryadLINQ
- 4 IBM Parallel Machine Learning Toolbox
- 5 Uniformly Fine-Grained Data-Parallel Computing for Machine Learning Algorithms
- Part Two Supervised and Unsupervised Learning Algorithms
- Part Three Alternative Learning Settings
- Part Four Applications
- Subject Index
- References
Summary
In many ways, the objective of the IBM Parallel Machine Learning Toolbox (PML) is similar to that of Google's MapReduce programming model (Dean and Ghemawat, 2004) and the open source Hadoop system, which is to provide Application Programming Interfaces (APIs) that enable programmers who have no prior experience in parallel and distributed systems to nevertheless implement parallel algorithms with relative ease. Like MapReduce and Hadoop, PML supports associative-commutative computations as its primary parallelization mechanism. Unlike MapReduce and Hadoop, PML fundamentally assumes that learning algorithms can be iterative in nature, requiring multiple passes over data. It also extends the associative-commutative computational model in various aspects, the most important of which are:
The ability to maintain the state of each worker node between iterations, making it possible, for example, to partition and distribute data structures across workers
Efficient distribution of data, including the ability for each worker to read a subset of the data, to sample the data, or to scan the entire dataset
Access to both sparse and dense datasets
Parallel merge operations using tree structures for efficient collection of worker results on very large clusters
In order to make these extensions to the computational model and still address ease of use, PML provides an object-oriented API in which algorithms are objects that implement a predefined set of interface methods. The PML infrastructure then uses these interface methods to distribute algorithm objects and their computations across multiple compute nodes.
Information
- Type
- Chapter
- Information
- Scaling up Machine LearningParallel and Distributed Approaches, pp. 69 - 88Publisher: Cambridge University PressPrint publication year: 2011
References
, and 1996. Parallel Mining of Association Rules. IEEE Transactions on Knowledge and Data Engineering.CrossRefGoogle Scholar
, and 1994. Fast Algorithms for Mining Association Rules. In: Proceedings of the International Conference on Very Large Data Bases (VLDB).Google Scholar
, and 1995. Mining Sequential Patterns. In: Proceedings of the International Conference on Data Engineering (ICDE).Google Scholar
, , and 1993. Mining Association Rules between Sets of Items in Large Databases. In: Proceedings of the International Conference on Management of Data (SIGMOD).Google Scholar
, , and 1998 (August). CLOUDS: Classification for Large or Out-of-Core Datasets. In: Conference on Knowledge Discovery and Data Mining.Google Scholar
, , , , , and 1999. Probabilistic Estimation Based Data Mining for Discovering Insurance Risks. IEEE Intelligent Systems, 14(6), 49–58.CrossRefGoogle Scholar
, , , , , , and 2001. Segmentation-BasedModeling for Advanced Targeted Marketing. Pages 408–413 of: Proceedings of the Seventh ACMSIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM.CrossRefGoogle Scholar
, , , and 2002. A Probabilistic Estimation Framework for Predictive Modeling Analytics. IBM Systems Journal, 41(3), 438–448.CrossRefGoogle Scholar
, and 2010. A Streaming Parallel Decision Tree Algorithm. Journal of Machine Learning Research, 11, 789–812.Google Scholar
, , and 1997. Beyond Market Basket: Generalizing Association Rules to Correlations. In: Proceedings of the International Conference on Management of Data (SIGMOD).Google Scholar
, and 2004. MapReduce: Simplified Data Processing on Large Clusters. In: Proceedings of the Symposium on Operating System Design and Implementation.Google Scholar
, and 1999. Efficient Mining of Emerging Patterns: Discovering Trends and Differences. In: Proceedings of the International Conference on Knowledge Discovery and Data Mining (SIGKDD).Google Scholar
, , , and 2006. Embedded Predictive Modeling in a Parallel Relational Database. Pages 569–574 of: SAC '06: Proceedings of the 2006 ACM Symposium on Applied Computing. New York: ACM.CrossRefGoogle Scholar
, , , and 1999 (June). BOAT – Optimistic Decision Tree Construction. Pages 169–180 of: ACM SIGMOD International Conference on Management of Data.CrossRefGoogle Scholar
, , and 1999. Efficient Mining of Partial Periodic Patterns in Time Series Database. In: Proceedings of the International Conference on Data Engineering (ICDE).Google Scholar
, and 2003 (May). Communication and Memory Efficient Parallel Decision Tree Construction. In: The 3rd SIAM International Conference on Data Mining.Google Scholar
, , and 1998 (March). ScalParC: A New Scalable and Efficient Parallel Classification Algorithm for Mining Large Datasets. Pages 573–579 of: The 12th International Parallel Processing Symposium.Google Scholar
, , and 1997. Discovery of Frequent Episodes in Event Sequences. Data Mining and Knowledge Discovery.Google Scholar
, , and 1996. SLIQ: A Fast Scalable Classifier for Data Mining. Pages 18–32 of: The 5th International Conference on Extending Database Technology.Google Scholar
, and 2001. Using Simulated Pseudo Data to Speed Up Statistical Predictive Modeling from Massive Data Sets. In: First SIAM International Conference on Data Mining.Google Scholar
, and 2002. Segmented Regression Estimators for Massive Data Sets. In: Second SIAM International Conference on Data Mining.Google Scholar
2006. Transform Regression and the Kolmogorov Superposition Theorem. In: Proceedings of the Sixth SIAM International Conference on Data Mining.Google Scholar
, and 2002. Leaning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. Cambridge, MA: MIT Press.Google Scholar
, , and 1996. SPRINT: A Scalable Parallel Classifier for DataMining. Pages 544–555 of: The 22nd International Conference on Very Large Databases.Google Scholar
, , , and 1998. Scalable Techniques for Mining Causal Structures. In: Proceedings of the International Conference on Very Large Data Bases (VLDB).Google Scholar
, , , and 2008. Pascal Large Scale Learning Challenge.
, , , and 1995. New Algorithms for Fast Discovery of Association Rules. In: Proceedings of the International Conference on Knowledge Discovery and Data Mining (SIGKDD).Google Scholar
, and 2002. A Large Scale Clustering Scheme for Kernel k-Means. Page 40289 of: Proceedings of the 16th International Conference on Pattern Recognition (ICPR'02) Volume 4. Washington, DC: IEEE Computer Society.Google Scholar
Accessibility standard: Unknown
Why this information is here
This section outlines the accessibility features of this content - including support for screen readers, full keyboard navigation and high-contrast display options. This may not be relevant for you.Accessibility Information
Accessibility compliance for the PDF of this chapter is currently unknown
and may be updated in the future.