�ݺ�ߣshows by User: adatao

�ݺ�ߣshows by User: adatao / http://www.slideshare.net/images/logo.gif �ݺ�ߣshows by User: adatao / Wed, 14 Oct 2015 12:12:37 GMT �ݺ�ߣShare feed for �ݺ�ߣshows by User: adatao First-ever scalable, distributed deep learning architecture using Spark & Tachyon /slideshow/firstever-scalable-distributed-deep-learning-architecture-using-spark-tachyon/53923191 stratanycfinal-151014121237-lva1-app6892
This talk was first presented at the 2015 Strata+Hadoop World NYC (http://strataconf.com/big-data-conference-ny-2015/public/schedule/detail/43484) Deep learning algorithms have been widely used in many real-world applications, including computer vision, machine translation, and fraud detection. Unfortunately, deep learning only works best when the model is big and trained on large-scale datasets. Meanwhile, distributed computing platforms like Spark are designed to handle big data, and have been used extensively. By having deep learning available on Spark, businesses can fully take advantage of deep learning capabilities on their datasets using their existing Spark infrastructure. In this talk, we present a scalable implementation of predictive deep learning algorithms on Spark, including feedforward neural networks, convolutional neural networks (CNNs), and recurrent neural networks (RNNs). This, to our best knowledge, is the first successful implementation of CNNs and RNNs on Spark. To support big model training, we use Tachyon as common storage layers between the Spark workers. With its in-memory distributed execution model, Tachyon provides a scalable approach even when the model is too big to be handled on a single machine. Our solution also exploits graphical processing units (GPUs) for matrix computation whenever they are available on worker nodes, further improving execution time. The attendees will learn about deep learning models, the architecture of the system, and how to train and run deep learning models on Spark with Tachyon.]]>
This talk was first presented at the 2015 Strata+Hadoop World NYC (http://strataconf.com/big-data-conference-ny-2015/public/schedule/detail/43484) Deep learning algorithms have been widely used in many real-world applications, including computer vision, machine translation, and fraud detection. Unfortunately, deep learning only works best when the model is big and trained on large-scale datasets. Meanwhile, distributed computing platforms like Spark are designed to handle big data, and have been used extensively. By having deep learning available on Spark, businesses can fully take advantage of deep learning capabilities on their datasets using their existing Spark infrastructure. In this talk, we present a scalable implementation of predictive deep learning algorithms on Spark, including feedforward neural networks, convolutional neural networks (CNNs), and recurrent neural networks (RNNs). This, to our best knowledge, is the first successful implementation of CNNs and RNNs on Spark. To support big model training, we use Tachyon as common storage layers between the Spark workers. With its in-memory distributed execution model, Tachyon provides a scalable approach even when the model is too big to be handled on a single machine. Our solution also exploits graphical processing units (GPUs) for matrix computation whenever they are available on worker nodes, further improving execution time. The attendees will learn about deep learning models, the architecture of the system, and how to train and run deep learning models on Spark with Tachyon.]]> Wed, 14 Oct 2015 12:12:37 GMT /slideshow/firstever-scalable-distributed-deep-learning-architecture-using-spark-tachyon/53923191 adatao@slideshare.net(adatao) First-ever scalable, distributed deep learning architecture using Spark & Tachyon adatao This talk was first presented at the 2015 Strata+Hadoop World NYC (http://strataconf.com/big-data-conference-ny-2015/public/schedule/detail/43484) Deep learning algorithms have been widely used in many real-world applications, including computer vision, machine translation, and fraud detection. Unfortunately, deep learning only works best when the model is big and trained on large-scale datasets. Meanwhile, distributed computing platforms like Spark are designed to handle big data, and have been used extensively. By having deep learning available on Spark, businesses can fully take advantage of deep learning capabilities on their datasets using their existing Spark infrastructure. In this talk, we present a scalable implementation of predictive deep learning algorithms on Spark, including feedforward neural networks, convolutional neural networks (CNNs), and recurrent neural networks (RNNs). This, to our best knowledge, is the first successful implementation of CNNs and RNNs on Spark. To support big model training, we use Tachyon as common storage layers between the Spark workers. With its in-memory distributed execution model, Tachyon provides a scalable approach even when the model is too big to be handled on a single machine. Our solution also exploits graphical processing units (GPUs) for matrix computation whenever they are available on worker nodes, further improving execution time. The attendees will learn about deep learning models, the architecture of the system, and how to train and run deep learning models on Spark with Tachyon. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/stratanycfinal-151014121237-lva1-app6892-thumbnail.jpg?width=120&height=120&fit=bounds" /> This talk was first presented at the 2015 Strata+Hadoop World NYC (http://strataconf.com/big-data-conference-ny-2015/public/schedule/detail/43484) Deep learning algorithms have been widely used in many real-world applications, including computer vision, machine translation, and fraud detection. Unfortunately, deep learning only works best when the model is big and trained on large-scale datasets. Meanwhile, distributed computing platforms like Spark are designed to handle big data, and have been used extensively. By having deep learning available on Spark, businesses can fully take advantage of deep learning capabilities on their datasets using their existing Spark infrastructure. In this talk, we present a scalable implementation of predictive deep learning algorithms on Spark, including feedforward neural networks, convolutional neural networks (CNNs), and recurrent neural networks (RNNs). This, to our best knowledge, is the first successful implementation of CNNs and RNNs on Spark. To support big model training, we use Tachyon as common storage layers between the Spark workers. With its in-memory distributed execution model, Tachyon provides a scalable approach even when the model is too big to be handled on a single machine. Our solution also exploits graphical processing units (GPUs) for matrix computation whenever they are available on worker nodes, further improving execution time. The attendees will learn about deep learning models, the architecture of the system, and how to train and run deep learning models on Spark with Tachyon.

First-ever scalable, distributed deep learning architecture using Spark & Tachyon from Arimo, Inc.

]]> 5766 7 https://cdn.slidesharecdn.com/ss_thumbnails/stratanycfinal-151014121237-lva1-app6892-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

1 Distributed DataFrame (DDF) Simplifying Big Data For The Rest Of Us /slideshow/distributed-dataframe-ddf-simplifying-big-data-for-the-rest-of-us/36849950 sparkddfr7-140710153126-phpapp01
]]>
]]> Thu, 10 Jul 2014 15:31:25 GMT /slideshow/distributed-dataframe-ddf-simplifying-big-data-for-the-rest-of-us/36849950 adatao@slideshare.net(adatao) Distributed DataFrame (DDF) Simplifying Big Data For The Rest Of Us adatao <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/sparkddfr7-140710153126-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" />

Distributed DataFrame (DDF) Simplifying Big Data For The Rest Of Us from Arimo, Inc.

]]> 1739 3 https://cdn.slidesharecdn.com/ss_thumbnails/sparkddfr7-140710153126-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

1 Adatao: Interactive, Visual, Predictive Analytics for Big Data @ Silicon Valley Machine Learning Meetup /slideshow/adatao-silicon-valley-machine-learning-meetup/30057801 2014-01-10adataosiliconvalleyml-140115141803-phpapp02
On Jan 10, 2014, we presented this to a full-house audience, members of the Silicon Valley Machine Learning group, at the Hacker Dojo in Mountain View, CA. The session was enjoyable with a very interactive, well-informed audience. ]]>
On Jan 10, 2014, we presented this to a full-house audience, members of the Silicon Valley Machine Learning group, at the Hacker Dojo in Mountain View, CA. The session was enjoyable with a very interactive, well-informed audience. ]]> Wed, 15 Jan 2014 14:18:03 GMT /slideshow/adatao-silicon-valley-machine-learning-meetup/30057801 adatao@slideshare.net(adatao) Adatao: Interactive, Visual, Predictive Analytics for Big Data @ Silicon Valley Machine Learning Meetup adatao On Jan 10, 2014, we presented this to a full-house audience, members of the Silicon Valley Machine Learning group, at the Hacker Dojo in Mountain View, CA. The session was enjoyable with a very interactive, well-informed audience. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/2014-01-10adataosiliconvalleyml-140115141803-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" /> On Jan 10, 2014, we presented this to a full-house audience, members of the Silicon Valley Machine Learning group, at the Hacker Dojo in Mountain View, CA. The session was enjoyable with a very interactive, well-informed audience.

Adatao: Interactive, Visual, Predictive Analytics for Big Data @ Silicon Valley Machine Learning Meetup from Arimo, Inc.

]]> 1916 4 https://cdn.slidesharecdn.com/ss_thumbnails/2014-01-10adataosiliconvalleyml-140115141803-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

1 Adatao Live Demo at the First Spark Summit /adatao/adatao-live-demo-at-the-first-spark-summit adataolivedemo-131213214640-phpapp02
Adatao's Live Product Demo at the First Spark Summit December 2, 2013 Nikko Hotel, San Francisco]]>
Adatao's Live Product Demo at the First Spark Summit December 2, 2013 Nikko Hotel, San Francisco]]> Fri, 13 Dec 2013 21:46:40 GMT /adatao/adatao-live-demo-at-the-first-spark-summit adatao@slideshare.net(adatao) Adatao Live Demo at the First Spark Summit adatao Adatao's Live Product Demo at the First Spark Summit December 2, 2013 Nikko Hotel, San Francisco <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/adataolivedemo-131213214640-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" /> Adatao's Live Product Demo at the First Spark Summit December 2, 2013 Nikko Hotel, San Francisco

Adatao Live Demo at the First Spark Summit from Arimo, Inc.

]]> 1867 4 https://cdn.slidesharecdn.com/ss_thumbnails/adataolivedemo-131213214640-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

1 Visual, Interactive, Predictive Analytics for Big Data /slideshow/adatao-spark-summit-demo/29057671 adatao-spark-summit-demo-131209202755-phpapp01
Adatao Demo at the First Apache Spark Summit, Nikko Hotel, San Francisco, December 2, 2013 A real-time, live demo of the Adatao big-data analytics system for both Business Users and Data Scientists/Engineers. We showed terabyte data modeling in seconds on a 40-node cluster. And with a beautiful, user-friendly web app, as well as R/R-Studio & Python interfaces.]]>
Adatao Demo at the First Apache Spark Summit, Nikko Hotel, San Francisco, December 2, 2013 A real-time, live demo of the Adatao big-data analytics system for both Business Users and Data Scientists/Engineers. We showed terabyte data modeling in seconds on a 40-node cluster. And with a beautiful, user-friendly web app, as well as R/R-Studio & Python interfaces.]]> Mon, 09 Dec 2013 20:27:55 GMT /slideshow/adatao-spark-summit-demo/29057671 adatao@slideshare.net(adatao) Visual, Interactive, Predictive Analytics for Big Data adatao Adatao Demo at the First Apache Spark Summit, Nikko Hotel, San Francisco, December 2, 2013 A real-time, live demo of the Adatao big-data analytics system for both Business Users and Data Scientists/Engineers. We showed terabyte data modeling in seconds on a 40-node cluster. And with a beautiful, user-friendly web app, as well as R/R-Studio & Python interfaces. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/adatao-spark-summit-demo-131209202755-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> Adatao Demo at the First Apache Spark Summit, Nikko Hotel, San Francisco, December 2, 2013 A real-time, live demo of the Adatao big-data analytics system for both Business Users and Data Scientists/Engineers. We showed terabyte data modeling in seconds on a 40-node cluster. And with a beautiful, user-friendly web app, as well as R/R-Studio & Python interfaces.

Visual, Interactive, Predictive Analytics for Big Data from Arimo, Inc.

]]> 1082 4 https://cdn.slidesharecdn.com/ss_thumbnails/adatao-spark-summit-demo-131209202755-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

1 Big Data, Big Opportunities /slideshow/adatao-keynote-address-uiuc-research-park-big-data-summit/29055841 adataokeynoteuiucresearchparkbigdatasummit-131209200325-phpapp02
Adatao Keynote Address @ UIUC Research Park Big-Data Summit, December 6, 2013 We were invited to give the Keynote address at the UIUC Research Park Big-Data Summit. We talked about (a) Why Big Data, (b) Big-Data Success Factors, and (c) The Future of Big Data. We also showed how Adatao approaches Big Data analysis for business users, via a beautiful, easy-to-use yet powerful, interactive web application.]]>
Adatao Keynote Address @ UIUC Research Park Big-Data Summit, December 6, 2013 We were invited to give the Keynote address at the UIUC Research Park Big-Data Summit. We talked about (a) Why Big Data, (b) Big-Data Success Factors, and (c) The Future of Big Data. We also showed how Adatao approaches Big Data analysis for business users, via a beautiful, easy-to-use yet powerful, interactive web application.]]> Mon, 09 Dec 2013 20:03:25 GMT /slideshow/adatao-keynote-address-uiuc-research-park-big-data-summit/29055841 adatao@slideshare.net(adatao) Big Data, Big Opportunities adatao Adatao Keynote Address @ UIUC Research Park Big-Data Summit, December 6, 2013 We were invited to give the Keynote address at the UIUC Research Park Big-Data Summit. We talked about (a) Why Big Data, (b) Big-Data Success Factors, and (c) The Future of Big Data. We also showed how Adatao approaches Big Data analysis for business users, via a beautiful, easy-to-use yet powerful, interactive web application. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/adataokeynoteuiucresearchparkbigdatasummit-131209200325-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" /> Adatao Keynote Address @ UIUC Research Park Big-Data Summit, December 6, 2013 We were invited to give the Keynote address at the UIUC Research Park Big-Data Summit. We talked about (a) Why Big Data, (b) Big-Data Success Factors, and (c) The Future of Big Data. We also showed how Adatao approaches Big Data analysis for business users, via a beautiful, easy-to-use yet powerful, interactive web application.

Big Data, Big Opportunities from Arimo, Inc.

]]> 1383 5 https://cdn.slidesharecdn.com/ss_thumbnails/adataokeynoteuiucresearchparkbigdatasummit-131209200325-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

1 https://cdn.slidesharecdn.com/profile-photo-adatao-48x48.jpg?cb=1528505961 Follow our new slideshare: http://www.slideshare.net/arimoinc ====== https://arimo.com arimo.com https://cdn.slidesharecdn.com/ss_thumbnails/stratanycfinal-151014121237-lva1-app6892-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/firstever-scalable-distributed-deep-learning-architecture-using-spark-tachyon/53923191 First-ever scalable, d... https://cdn.slidesharecdn.com/ss_thumbnails/sparkddfr7-140710153126-phpapp01-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/distributed-dataframe-ddf-simplifying-big-data-for-the-rest-of-us/36849950 Distributed DataFrame ... https://cdn.slidesharecdn.com/ss_thumbnails/2014-01-10adataosiliconvalleyml-140115141803-phpapp02-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/adatao-silicon-valley-machine-learning-meetup/30057801 Adatao: Interactive, V...