[Palo Alto] TensorFlow +Spark +Neural Nets +Deep Learning +Nvidia CUDA +OpenDeep

This is a past event

395 people went

Location image of event venue

Details

*** Note the Earlier Start Time: 5:30pm ***

This event will be recorded for off-line viewing.

Parking and Caltrain Details

=================

There is a free parking structure with plenty of parking available across the street at 261 Sheridan Ave (https://www.google.com/maps/place/261+Sheridan+Ave,+Palo+Alto,+CA+94306/data=!4m2!3m1!1s0x808fbaf01defdecf:0x65a6ca10f23a68ba?sa=X&ved=0ahUKEwjc-OePtfrLAhUS5GMKHXILDEQQ8gEIHDAA). There is also overflow parking lots two blocks away at the intersection of Birch St. and Sherman Ave (https://www.google.com/maps/place/2498+Birch+St,+Palo+Alto,+CA+94306/@37.4268638,-122.1434993,19z/data=!3m1!4b1!4m2!3m1!1s0x808fbae576f28143:0x90816dc74c26e245).

For those taking the CalTrain, we are a 5-minute walk from the California Ave stop.

=================

Agenda

5:30pm: Arrive and Mingle

6:00pm: Introductions and Exciting Announcements

6:30pm: Talk 1

Modularity in Neural Nets and Resulting Design Choices in Open Deep (http://www.opendeep.org/)

Markus Beissinger (https://www.linkedin.com/in/mbeissinger), Founder of Vitruvian Science Labs (http://vitruvianscience.com/)

7:00pm: Talk 2

DeepLearning4J (http://deeplearning4j.org/): CUDA-based Deep Learning Java Library for Spark

Adam Gibson (https://www.linkedin.com/in/agibsonccc), Founder of Skymind (http://skymind.io)

7:30pm: Talk 3

Spark Project Tungsten + GPUs (http://www.slideshare.net/ishizaki/exploiting-gpus-in-spark): Exploiting GPUs in Spark

Kazuaki Ishizaki (https://www.linkedin.com/in/kazuaki-ishizaki-63744113), Research Engineer, IBM Research - Tokyo

8:00pm: Talk 4

Nvidia CUDA (https://developer.nvidia.com/cuda-zone) + GPUs + Spark: Extending Spark Operators for Distributed Spark Matrix Multiplication in new, Row-grouped CSR Format (http://arxiv.org/pdf/1012.2270.pdf) for sparse matrices

Maxim Naumov (https://www.linkedin.com/in/maxim-naumov-86133250), Sr. Research Scientist @ Nvidia (http://nvidia.com/)

8:30pm: Talk 5

TensorFlow (https://www.tensorflow.org/) + Scikit-learn (http://scikit-learn.org/stable/) + TensorFlow Serving (https://www.tensorflow.org/versions/master/tutorials/tfserve/index.html) + GPUs

Chris Fregly (http://linkedin.com/in/cfregly), Principal Data Solutions Engineer, IBM Spark Tech Center (http://spark.tc)

9:00pm: Go Home and Spend Time With Your Family!

Related Links

http://www.slideshare.net/ishizaki/exploiting-gpus-in-spark

https://developer.nvidia.com/cuda-zone

http://www.opendeep.org/

https://databricks.com/blog/2016/01/25/deep-learning-with-spark-and-tensorflow.html (https://github.com/amplab/SparkNet/)

https://github.com/amplab/SparkNet

http://arxiv.org/pdf/1511.06051v3.pdf

http://deeplearning4j.org/

http://ampcamp.berkeley.edu/5/exercises/image-classification-with-pipelines.html

https://github.com/apache/spark/blob/master/sql/core/src/main/scala/org/apache/spark/sql/execution/basicOperators.scala

https://github.com/apache/spark/blob/master/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/logical/basicOperators.scala

http://www.datascienceassn.org/content/spark-gets-gpu-lab

http://www.datascienceassn.org/content/single-gpu-powered-node-4x-faster-50-node-spark-cluster

https://issues.apache.org/jira/browse/SPARK-3785

https://issues.apache.org/jira/browse/SPARK-3434

https://groups.google.com/forum/#!topic/scalanlp-discuss/ZZJbXtwL19c

https://github.com/BIDData

David Hall: https://www.linkedin.com/profile/view?id=ADEAAABq7i4B_GWYFbBSB6KIL2DEMYCGfiHUCOo

Tungsten + GPUs + UnsafeColumn: http://www.slideshare.net/SparkSummit/exploiting-gpus-for-columnar-datafrrames-by-kiran-lonikar

Wikipedia (for laughs):

https://en.wikipedia.org/wiki/Deep_learning