Slideshare uses cookies to improve functionality and performance, and to provide you with relevant advertising. If you continue browsing the site, you agree to the use of cookies on this website. See our User Agreement and Privacy Policy.

Slideshare uses cookies to improve functionality and performance, and to provide you with relevant advertising. If you continue browsing the site, you agree to the use of cookies on this website. See our Privacy Policy and User Agreement for details.

Successfully reported this slideshow.

Like this presentation? Why not share!

- Deep Learning for NLP (without Magi... by BigDataCloud 7410 views
- RBM with DL4J for Deep Learning by 신동 강 1568 views
- Deep learning on a mixed cluster wi... by François Garillot 507 views
- Deep Learning for Java (DL4J) by 신동 강 4219 views
- Neural Networks and Deep Learning by Asim Jalis 2982 views
- Distributed Deep Learning on Hadoop... by DataWorks Summit/... 6015 views

2,739 views

Published on

No Downloads

Total views

2,739

On SlideShare

0

From Embeds

0

Number of Embeds

69

Shares

0

Downloads

0

Comments

0

Likes

8

No embeds

No notes for slide

Acyclic data flow is a powerful abstraction, but is not efficient for applications that repeatedly reuse a working set of data:

Iterative algorithms (many in machine learning)

• No single programming model or framework can excel at

every problem; there are always tradeoﬀs between simplicity, expressivity, fault tolerance, performance, etc.

Talking points:

wanted to start with a known tool to the hadoop community, with expected characteristics

Mahout’s SGD is well known, and so we used that as a base point

- 1. Adam Gibson Email: 0@blix.io Twitter: @agibsonccc Github: https://github.com/a gibsonccc Slideshare: http://slideshare.net/agibsonc cc/ Instructor at http://zipfianacademy.com/ Wired Coverage: http://www.wired.com/2014/0 6/skymind-deep-learning/
- 2. Josh Patterson Email: josh@pattersonconsultingtn.com Twitter: @jpatanooga Github: https://github.com/jpata nooga Past Published in IAAI-09: “TinyTermite: A Secure Routing Algorithm” Grad work in Meta-heuristics, Ant-algorithms Tennessee Valley Authority (TVA) Hadoop and the Smartgrid Cloudera Principal Solution Architect Today: Patterson Consulting
- 3. Overview • What is Deep Learning? • Deep Belief Networks • Implementation on Hadoop/YARN • Results
- 4. What is Deep Learning? Algorithm that tries to learn simple features in lower layers And more complex features in higher layers
- 5. Interesting Properties of Deep Learning Reduces a problem with overfitting in neural networks. Introduces new techniques for "unsupervised feature learning” introduces new more automatic ways to figure out the parts of your data you should feed into your learning algorithm.
- 6. Chasing Nature Learning sparse representations of auditory signals leads to filters that closely correspond to neurons in early audio processing in mammals When applied to speech Learned representations showed a striking resemblance to the cochlear filters in the auditory cortext
- 7. Yann LeCunn on Deep Learning Has become the dominant method for acoustic modeling in speech recognition Quickly becoming the dominant method for several vision tasks such as object recognition object detection semantic segmentation.
- 8. What is a Deep Belief Network? Generative probabilistic model Composed of one visible layer Many hidden layers Each hidden layer learns relationship between units in lower layer Higher layer representations tend to become more complext
- 9. Restricted Boltzmann Machines Unsupervised model: Does feature learning by repeated sampling of the input data. Learns how to reconstruct data for good feature detection. RBMs have different formulas for different kinds of data: Binary Continuous
- 10. DeepLearning4J Implementation in Java Self-contained & built on Akka, Hazelcast, Jblas Distributed to run faster and with more features than current Theano-based implementations. Talks to any data source, expects one format.
- 11. Vectorized Implementation Handles lots of data concurrently. Any number of examples at once, but the code does not change. Faster: Allows for native/GPU execution. One format: Everything is a matrix.
- 12. DL4J vs Theano Perf GPUs are inherently faster than normal native. Theano is not distributed, and GPUs have very low RAM. DL4J allows for situations where you have to “throw CPUs at it.”
- 13. What are Good Applications for Deep Learning? Image Processing High MNIST Scores Audio Processing Current Champ on TIMIT dataset Text / NLP Processing Word2vec, etc
- 14. Past Work: Parallel Iterative Algorithms on YARN Started with Parallel linear, logistic regression Parallel Neural Networks Packaged in Metronome 100% Java, ASF 2.0 Licensed, on github
- 15. Parameter Averaging McDonald, 2010 Distributed Training Strategies for the Structured Perceptron Langford, 2007 Vowpal Wabbit Jeff Dean’s Work on Parallel SGD DownPour SGD 19
- 16. MapReduce vs. Parallel Iterative 20 Input Output Map Map Map Reduce Reduce Processor Processor Processor Superstep 1 Processor Processor Superstep 2 . . . Processor
- 17. SGD: Serial vs Parallel 21 Model Training Data Worker 1 Master Partial Model Global Model Worker 2 Partial Model Worker N Partial Model Split 1 Split 2 Split 3 …
- 18. Managing Resources Running through YARN on hadoop is important Allows for workflow scheduling Allows for scheduler oversight Allows the jobs to be first class citizens on Hadoop And share resources nicely
- 19. Parallelizing Deep Belief Networks Two phase training Pre Train Fine tune Each phase can do multiple passes over dataset Entire network is averaged at master
- 20. PreTrain and Lots of Data We’re exploring how to better leverage the unsupervised aspects of the PreTrain phase of Deep Belief Networks Allows for the use of far less unlabeled data Allows us to more easily modeled the massive amounts of structured data in HDFS
- 21. DBNs on IR Performance Faster to Train. Parameter averaging is an automatic form of regularization. Adagrad with IR allows for better generalization of different features and even pacing.
- 22. Scale Out Metrics Batches of records can be processed by as many workers as there are data splits Message passing overhead is minimal Exhibits linear scaling Example: 3x workers, 3x faster learning
- 23. Usage From Command Line Run Deep Learning on Hadoop yarn jar iterativereduce-0.1-SNAPSHOT.jar [props file] Evaluate model ./score_model.sh [props file]
- 24. Handwriting Renders
- 25. Faces Renders
- 26. …In Which We Gather Lots of Cat Photos
- 27. Future Direction GPUs Better Vectorization tooling Move YARN version back over to JBLAS for matrices
- 28. References “A Fast Learning Algorithm for Deep Belief Nets” Hinton, G. E., Osindero, S. and Teh, Y. - Neural Computation (2006) “Large Scale Distributed Deep Networks” Dean, Corrado, Monga - NIPS (2012) “Visually Debugging Restricted Boltzmann Machine Training with a 3D Example” Yosinski, Lipson - Representation Learning Workshop (2012)

No public clipboards found for this slide

Be the first to comment