Scipediacontent: Scipediacontent moved page Draft Content 125564836 to Ahmed et al 2019a

2021-01-28T16:28:12Z

Scipediacontent moved page Draft Content 125564836 to Ahmed et al 2019a

Scipediacontent: Created page with " == Abstract == Machine Learning is transitioning from an art and science into a technology available to every developer. In the near future, every application on every platf..."

2021-01-28T16:28:09Z

Created page with " == Abstract == Machine Learning is transitioning from an art and science into a technology available to every developer. In the near future, every application on every platf..."

New page

== Abstract ==

Machine Learning is transitioning from an art and science into a technology available to every developer. In the near future, every application on every platform will incorporate trained models to encode data-based decisions that would be impossible for developers to author. This presents a significant engineering challenge, since currently data science and modeling are largely decoupled from standard software development processes. This separation makes incorporating machine learning capabilities inside applications unnecessarily costly and difficult, and furthermore discourage developers from embracing ML in first place. In this paper we present ML .NET, a framework developed at Microsoft over the last decade in response to the challenge of making it easy to ship machine learning models in large software applications. We present its architecture, and illuminate the application demands that shaped it. Specifically, we introduce DataView, the core data abstraction of ML .NET which allows it to capture full predictive pipelines efficiently and consistently across training and inference lifecycles. We close the paper with a surprisingly favorable performance study of ML .NET compared to more recent entrants, and a discussion of some lessons learned.

== Original document ==

The different versions of the original document can be found in:

* [http://arxiv.org/abs/1905.05715 http://arxiv.org/abs/1905.05715]

* [http://arxiv.org/pdf/1905.05715 http://arxiv.org/pdf/1905.05715]

* [http://dl.acm.org/ft_gateway.cfm?id=3330667&ftid=2075615&dwn=1 http://dl.acm.org/ft_gateway.cfm?id=3330667&ftid=2075615&dwn=1],
: [http://dx.doi.org/10.1145/3292500.3330667 http://dx.doi.org/10.1145/3292500.3330667] under the license http://www.acm.org/publications/policies/copyright_policy#Background

* [https://dblp.uni-trier.de/db/conf/kdd/kdd2019.html#AhmedABCCDDEFFG19 https://dblp.uni-trier.de/db/conf/kdd/kdd2019.html#AhmedABCCDDEFFG19],
: [https://arxiv.org/pdf/1905.05715.pdf https://arxiv.org/pdf/1905.05715.pdf],
: [https://openreview.net/pdf?id=HJgt13-t97 https://openreview.net/pdf?id=HJgt13-t97],
: [https://arxiv.org/abs/1905.05715 https://arxiv.org/abs/1905.05715],
: [http://www.arxiv-vanity.com/papers/1905.05715 http://www.arxiv-vanity.com/papers/1905.05715],
: [https://openreview.net/forum?id=HJgt13-t97 https://openreview.net/forum?id=HJgt13-t97],
: [https://dl.acm.org/ft_gateway.cfm?id=3330667&ftid=2075615&dwn=1 https://dl.acm.org/ft_gateway.cfm?id=3330667&ftid=2075615&dwn=1],
: [https://au.arxiv.org/abs/1905.05715 https://au.arxiv.org/abs/1905.05715],
: [https://il.arxiv.org/abs/1905.05715 https://il.arxiv.org/abs/1905.05715],
: [https://uk.arxiv.org/abs/1905.05715 https://uk.arxiv.org/abs/1905.05715],
: [https://export.arxiv.org/abs/1905.05715 https://export.arxiv.org/abs/1905.05715],
: [http://export.arxiv.org/pdf/1905.05715 http://export.arxiv.org/pdf/1905.05715],
: [https://academic.microsoft.com/#/detail/2922522433 https://academic.microsoft.com/#/detail/2922522433]

← Older revision	Revision as of 16:28, 28 January 2021
(No difference)

Ahmed et al 2019a - Revision history

Scipediacontent: Scipediacontent moved page Draft Content 125564836 to Ahmed et al 2019a

Scipediacontent: Created page with " == Abstract == Machine Learning is transitioning from an art and science into a technology available to every developer. In the near future, every application on every platf..."