Skymind

Skymind
Privately held
Industry Analytics, Artificial Intelligence
Founders Adam Gibson, Chris Nicholson
Headquarters San Francisco, California, U.S., Japan
Area served
 Worldwide
Products Deeplearning4j, ND4J
Number of employees
 10-20 [1]
Website skymind.io

Skymind is a machine intelligence company supporting the open source deep learning framework Deeplearning4j and the JVM-based scientific computing library ND4J. The company was founded in 2014 by Adam Gibson and Chris Nicholson, and is headquartered in San Francisco, California. It is privately funded, and raised $3 million in seed financing in 2016.[2] [3]

Technology

Skymind has implemented neural networks such as convolutional networks, recurrent networks (including LSTMs), restricted Boltzmann machines, deep belief networks, deep autoencoders and word2vec.

Applications and Use Cases

Skymind's deep neural networks can be applied to use cases such as fraud and anomaly detection, recommender systems, machine vision, machine translation, machine transcription, face and voice recognition, time series predictions, business intelligence and econometric analytics. They are able to perform dimensionality reduction, classification, regression, collaborative filtering, feature learning and topic modeling.[4]

Skymind Intelligence Layer (SKIL)

The Skymind Intelligence Layer is Skymind's enterprise distribution of its open-source libraries. It is a pre-built and pre-configured collection of software packages focused on machine learning that can be installed and used together. SKIL is an operating system for deep learning, integrating with Hadoop, Spark and Kafka while running dockerized on Mesos.[5]

Deeplearning4j

Deeplearning4j is an open source deep learning library written for Java and the Java Virtual Machine[6][7] and a computing framework with wide support for deep learning algorithms. Its algorithms all include distributed parallel versions that integrate with Hadoop and Spark.[8]

ND4J

ND4J is a free, open-source extension of the Java programming language operating on the Java Virtual Machine—though it is compatible with both Scala and Clojure.[9] It includes the Scala wrapper ND4S.[10]

DataVec

DataVec is a software library for the preprocessing and vectorization of data. DataVec can normalize, standardize and vectorize all major data types, such as text, time series, image, video and sound. [11]

JavaCPP

JavaCPP is the bridge between the Java and C++ programming languages, which avoids much of the overhead of the Java Native Interface (JNI) to allow programmers on the JVM a way to perform computationally intensive tasks quickly.[12] It was created and is maintained by a Skymind engineer.

References

  1. ↑ https://www.crunchbase.com/organization/skymind
  2. ↑ "Skymind raises $3M to bring its Java deep-learning library to the masses". 2016-09-28. Retrieved 2016-10-21.
  3. ↑ "Blood, Sweat and Years: Raising Money As A Deep Learning Startup". 2016-10-23. Retrieved 2016-10-24.
  4. ↑ "Interview with the man who's bringing scalable, deep learning to businesses". 2016-10-24. Retrieved 2016-10-24.
  5. ↑ "Deep learning startup Skymind raises $3 million, launches Intelligence Layer distribution". VentureBeat. 2016-09-28. Retrieved 2016-10-21.
  6. ↑ Metz, Cade (2014-06-02). "The Mission to Bring Google's AI to the Rest of the World". Wired.com. Retrieved 2014-06-28.
  7. ↑ Vance, Ashlee (2014-06-03). "Deep Learning for (Some of) the People". Bloomberg Businessweek. Retrieved 2014-06-28.
  8. ↑ "Adam Gibson, DeepLearning4j on Spark and Data Science on JVM with nd4j, SF Spark @Galvanize 20150212". SF Spark Meetup. 2015-02-12. Retrieved 2015-03-01.
  9. ↑ "ND4J: Scientific Computing for Java & Scala".
  10. ↑ "Github Repository".
  11. ↑ https://github.com/deeplearning4j/datavec
  12. ↑ https://github.com/bytedeco/javacpp
This article is issued from Wikipedia - version of the 10/29/2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.