NML 2.0 AI Programming Language
April 9, 2018
the rising demand, most computer programmers and data scientists lack
the specialized knowledge and tools required to build deep learning
software solutions for their organizations. To address this,
DimensionalMechanics has launched its next-gen NeoPulse Framework 2.0,
and an easy-to-use programming language, called NML 2.0, to make
artificial intelligence (AI) more accessible to any software developer.
The company also announced it closed its A-2 round with $1,249k. The
Company converted $4,220 of debt to Preferred Stock and closed on an
incremental $1,249k raise for a total round of $5,469k Friday, April 6th
bringing the total raised to $7.9 million.
NeoPulse Framework 2.0 enables organizations to solve the most
complicated AI challenges at a fraction of the time and cost using
current methods or developing their own solutions in-house. Many
organizations have vast quantities of structured and unstructured data,
but lack the capabilities to convert the data into organizational assets
or intelligence. Leveraging NeoPulse Framework 2.0 and
DimensionalMechanics methodologies, organizations can build
sophisticated AI solutions based on all data types – video, numerical,
text, images or audio.
According to Gartner's "2018 Will Mark the Beginning of AI
Democratization," a survey of 3,160 CIOs view making progress with AI
initiatives as one of their top-five priorities for 2018. "In 2018,
organizations will strive to improve their understanding of what AI is
best suited to, and how to deploy it," says Chirag Dekate, research
director at Gartner. "By 2020, 85% of CIOs will be piloting AI programs
through a combination of buy, build and outsource efforts." The post
goes on to say, "However, CIOs will have to overcome challenges. Many
are dealing with data of poor or uncertain quality. Their organizations
often have minimal AI skills. Some CIOs are struggling to understand the
capabilities of new AI techniques, and how to identify use cases to
which AI may be applied productively."*
"With NeoPulse, our AI builds your AI," said Rajeev Dutt, CEO of
DimensionalMechanics. "We've created a platform for AI development that
lets developers of all skill levels rapidly answer data-driven questions
using a cost-effective and repeatable approach. Our NML language is
purpose-built for deep learning, yet easy enough to learn in just days.
Developers can now create AI models in a fraction of the time and cost
than was ever possible before."
Dutt continued, "Our versatile platform and software tools can be used
across any industry and any data types, and provides CIOs with a common
AI platform for use throughout their organizations. We're confident that
our next-gen NeoPulse can help accelerate the adoption of machine
learning for organizations of all sizes and technical capabilities."
NeoPulse Framework 2.0: Simplifying AI
NeoPulse Framework 2.0 gives developers the power to create custom AI
models to answer real-world business questions, based on their own data,
without being machine learning (ML) experts themselves. And, for those
with ML expertise, it can dramatically reduce the amount of code
required to create AI models by up to 85%. An animated video that shows
how it all works is viewable here. NeoPulse Framework 2.0 consists of
four main elements:
Modeling Language 2.0 (NML) is an intuitive DSL (domain specific
language) that fully automates the creation of new AI models to tackle
common ML problems (e.g. classification, regression, etc.) regardless of
data type (e.g. video, audio, images, etc.). The language is
Turing-complete and can accomplish in dozens of line of code what in
conventional languages (e.g. Python) would take hundreds of lines of
code. A programmer can learn NML in a matter of days with a simple
command line interface that is executed by NeoPulse AI Studio, and add
their own ML algorithms.
•NeoPulse AI Studio 2.0 is an AI that
builds AI. This AI-as-a-service is a server-based application, powered
by its own robust AI, dubbed "the oracle," that helps automate the
process of creating custom AI models. The most significant advancement
in the new NeoPulse Framework 2.0 is the addition of next-gen
hyperparameter optimization, which significantly outperforms
conventional Bayesian optimization approaches. Instead of relying on ML
experts for the difficult tasks of identifying key parameters needed for
data analysis, AI Studio's "oracle" is able to set those hyperparameters
when programmers either do not know how, or are not inclined to do so.
This advancement also means that AI Studio is typically able to
outperform ML experts attempting to solve the same problem.
•Portable Inference Models (PIMs) are
portable AI models, created by NeoPulse AI Studio 2.0 that can be
deployed across different computers, either locally on-premise or in the
cloud. The files are containerized artificial neural networks that can
be queried using a runtime layer.
•NeoPulse Query Runtime 2.0 (NPQR) is
a simple program that allows any application to access the AI models (PIMs)
using web-based REST APIs.
DimensionalMechanics gives organizations the versatility to have their
AI solutions in the cloud or on-premise. NeoPulse Framework 2.0 is
available via cloud deployment, with NeoPulse AI Studio and NeoPulse
Query Runtime, downloadable within Amazon Web Services (AWS)
Marketplace. An on-premise version of NeoPulse Framework 2.0 is expected
to be available later this year.
"DM on AWS was easy to implement and works flawlessly. Their technology
integrated right into our existing data and AI pipelines which is often
a major barrier to an AI strategy," said Tony Kippen, VP of Business