Cloud Ace, Datatonic, Deloitte Consulting, Devoteam, Pandera Systems, Pythian Services and Quantiphi are among the 50-plus Google Cloud partners with employees who’ve already earned the cloud provider’s new Professional Machine Learning Engineer certification.

The certification, unveiled last week, validates cloud professionals’ expertise in designing, building and productionizing machine-learning (ML) models to solve business challenges using Google Cloud technologies, along with their knowledge of proven ML models and techniques.

Finding employees with the right ML skills has been among the top challenges for IT leaders this year, Google Cloud said.

The pre-pandemic tech talent shortage challenged many organizations’ digital transformations, and they’re now playing catch-up, according to a September report from staffing company Robert Half International that highlights IT industry trends and starting salaries that its recruiters expect to see next year.

“They are faced with the need to accelerate their transformation process as well as address technical debt within their organization,” the report said. “Many are seeking technology professionals with expertise in AI and machine learning, cloud computing and robotic process automation.”

Artificial intelligence (AI)/ML specialists are expected to have among the highest median starting salaries for non-executive IT jobs in the U.S. next year, according to the report.

When Ottawa’s Pythian Services heard earlier this year there potentially would be a new ML-related certification coming from Google Cloud, it put its ML team on notice and had them stand by to get it done, according to Vanessa Simmons, vice president of business development for the cloud, data and analytics services company, a Google Cloud Premier Partner.

Pythian currently has two Google Cloud-certified Professional Machine Learning Engineers, with the goal of having its entire team certified. Carlos Timoteo, a Pythian data scientist and machine- learning engineer, has been working with Google Cloud, applying AI Platform, BigQuery and big data services to implement ML solutions for the company’s customers.

Timoteo took his first Google Cloud certification exam, the Google Cloud Professional Data Engineer, in 2017. Since then, he and his work colleagues and fellow data scientists in his circle have been waiting for a well-designed data scientist/ML certification, he said.

“The preparation was not too hard or long given my experience as a data scientist leveraging Google Cloud,” Timoteo said. “I used the provided preparation guide to identify the points I needed to study more to ace the exam, leveraging Google Cloud documentation, the Google Developer Machine Learning Crash Course and a couple books.”

The Google Cloud Professional Machine Learning Engineer certification exam has a big emphasis on engineering ML solutions, according to Timoteo. The data science portion of the exam is more focused on the technique than on the algorithm details, implementation and limitation, he said. Beginners should expect snippets of code in Python and SQL and should learn TensorFlow 2.x and its ecosystem and how to implement TensorFlow 2 on Google Cloud Platform (GCP) in production, he noted.

“The exam is a valuable tool in assessing if the exam taker is able to propose a solution that satisfies many requirements recurrent for a variety of solutions, in several industry verticals,” Timoteo said. “With this added experience, Google and our customers can trust in my team and myself to build the most elegant and suitable solution to satisfy their business demands leveraging Google products and best practices in the market.”

London-based Datatonic, Google Cloud’s 2019 Specialization Partner of the Year for AI and ML, has two employees who earned the new certification.

“The Google Cloud Professional Machine Learning Engineer certificate gives a valuable overview of production ML on Google Cloud Platform, particularly on designing solutions with ML best practices in mind, such as mitigating model bias and utilizing GCP tools to interpret model predictions,” said Julian West, a Datatonic data scientist. “This will be actionable for future projects with clients increasingly conscious of model bias and explainability in their ML projects.”

Three Deloitte Consulting Google Cloud practitioners have earned the new certification.

“Deloitte teamed with Google early on with unwavering commitment to the certification program given their market leadership in AI/ML,” said Tom Galizia, lead commercial partner for Deloitte Consulting‘s Alphabet/Google alliance. “[Google CEO Sundar Pichai] has stated Alphabet overall is an AI-first company, which is clearly reinforced with their broad and deep portfolio of AI/ML/analytics-based technologies at unprecedented cost curves and commitment to the democratization of AI/ML.”

The Google Cloud Professional Machine Learning Engineer certification requires a two-hour exam. The cloud provider recommends candidates have at least three years of industry experience, including one or more years designing and managing solutions using GCP. An ML engineer is proficient in model architecture, data pipeline interaction and metrics interpretation, and requires familiarity with application development, infrastructure management, data engineering and security, according to Google Cloud.

The certification exam evaluates candidates’ abilities to frame ML problems, develop ML models and architect ML solutions. It also assesses their abilities to automate and orchestrate ML pipelines, prepare and process data, and monitor, optimize and maintain ML solutions.

Google Cloud partners also can earn Expertises in AI and ML, including in Google Cloud AI and ML APIs, Contact Center AI, Document AI and Visual Intelligence. Quantiphi, SoftServe and SpringML are among the more than 90 partners with those Expertises.

Partners who earn a Google Cloud Specialization in Machine Learning signal the strongest level of Google Cloud ML proficiency and experience. Accenture, Atos, Deloitte Consulting, DoiT International, Quanitphi and Pythian have achieved those designations.

The post Google Cloud Debuts Professional Machine Learning Engineer Certification appeared first on Artificial Intelligence.

Internet of Things Inc (CVE:ITT) (OTCMKTS:INOTF) has revealed that its Weather Telematics Inc subsidiary is collaborating with IBI Group (TSE:IBG) to develop the Winter Ice and Snow Decision Support System for Operations Management (WISDOM) platform.

WISDOM is a resource management and decision support tool that uses artificial intelligence to assist in real-time responses to winter snow/ice road clearing operations. 

In a statement Thursday, Internet of Things said Weather Telematics and IBI have partnered with two Ontario counties to pilot the platform in the winter of 2020-21. They were brought together by the Ontario Centres of Excellence. 

Winter road maintenance accounts for a large portion of Canadian cities’ operating expenses, the company noted.  In 2018 alone, governments across North America spent upwards of $2 billion on winter operations.

With rapid population growth and more frequent extreme weather events, the costs of snow clearing are projected to increase further each year, prompting governments to analyze means to improve winter maintenance efficiency.

“We are extremely excited to be collaborating with IBI Group on the WISDOM platform,” Internet of Things CEO Michael Lende said in a statement. 

“Surprisingly, most governments rely on basic solutions to manage their winter maintenance and cleaning processes. WISDOM has the potential of dramatically improving the ability to manage government resources, significantly lower costs and improve driver safety.”

Derek Sims, IBI’s global director of intelligence said the platform “holds great promise in its ability to help governments better manage their winter snow/ice road clearing operations, and to lower their use of salt products without negatively impacting vehicle and pedestrian safety.”

The WISDOM platform consists of three interactive components:

• Real-time traffic and AI-predicted road conditions information which is used to route snowplows to designated areas efficiently, while predicted road conditions allow for more informed resource deployment in response to upcoming weather events
• Automation for the dynamic routing of appropriate vehicles to assist dispatchers
• Deployment of sensors on select vehicles to obtain ground-truth road condition data that feeds into the previous two components, bettering prediction accuracy and adjusting responses

Together, these components create a feedback loop that achieves precise dispatching of winter operation units and reduces excessive use of salt and de-icing materials, Internet of Things said.

The post Internet of Things says its Weather Telematics unit developing AI platform to assist in winter road clearing operations appeared first on Artificial Intelligence.

Thanks to smartphones and their downsized keyboards, autocomplete has become a nearly ubiquitous feature of how we write these days. To save us precious seconds composing and (at least in my fat-thumbed case) correcting words, our keyboards now prompt us with suggestions of what we’re trying to write to get the job done a little bit more easily. But email and messaging composing isn’t the only area where artificial intelligence and semantic analytics are being used in this way. Today, a startup that has built a platform that applies the concept to the world of coding is announcing a round of funding to expand its business.

Codota, an Israeli startup that provides an AI tool to developers to let them autocomplete strings of code that they are writing — intended both to speed up their work (it claims to “boost productivity by 25%”) and to make sure that it’s using the right syntax and ‘spelled’ correctly — has picked up $12 million, a Series A led by e.ventures, with participation also from previous backer Khosla Ventures, along with new investors TPY Capital and Hetz Ventures. The company has now raised $16 million in total, and it’s not disclosing its valuation.

The funding comes on the heels of Codota  acquiring one of its bigger competitors, TabNine, out of Canada, late last year (announced only in March however) to expand the number of languages that it can support. It now says it supports all major languages, including Python, JavaScript, Java, C, and HTML; and it operates across a major integrated development environments such as VSCode, Eclipse, and IntelliJ.

The funding will be used to expand its reach into that existing range further, as well as to bring on more customers. Today, the list of those that are already using Codota’s tools is impressive. It includes developers from companies like Google and Amazon, as well as Netflix, Alibaba, Airbnb and Atlassian, amongst many others. It says its user base has grown more than 1,000% in the last year, number over 1 million developers using it monthly.

The funding news coincides with Codota launching a new version of its autocompletion for JavaScript that merges Codota’s semantic technology with TabNine’s textual tech.

The first two names in the customer list above list are particularly surprising, but they underscore what Codota has focused on and seems to be doing right. These two tech giants are AI powerhouses in their own rights; both build and ship formidable sets of tools for developers; and Google specifically is one of the names most synonymous with autocomplete by way of the tools it’s built for Gmail.

The reason Codota — which was founded in 2015 — has had traction, said co-founder and CTO Eran Yahav, is because in fact coding has been a tough nut to crack for semantics teams, despite their advances in other languages.

“Up until a couple of years ago it wasn’t feasible,” he said, noting that four streams of technology are coming together when building auto-completion for coding: high-quality open source code availability to feed the algorithms; advances in semantic analysis to extract insights at scale; machine learning advances that essentially bring ML costs down; and computational resources to run everything in the cloud to make it something everyone can use everywhere. With open source really booming, and everything else coming along, it’s been a perfect storm that Codota has seized, even as others are working on this too.

“Others have done this with varying degrees of success,” Dror Weiss, the other co-founder and CEO said. “I’m guessing and also know that others are doing the same.” Others include the likes of Kite, Ubisoft and Mozilla, and a number of others.

One aspect that Codota has been building that is particularly timely now is the ability not just to provide more precise coding assistance to developers, but to “learn” what is best practice in a particular environment or workplace (it offers both individual and enterprise tiers, and it’s the latter that provides this feature). This can be useful in any situation, but particularly today when developers are working at home and on their own, giving them instant help as if they were physically in the same space.

Notably, while so much AI seems bent on the idea of autonomous systems, Weiss said emphatically that this is not the intention short or even long-term.

“I don’t think we would be able to replace developers and I don’t think we would want to,” he said. “Our aim is to take the mundane and repetitive aspects and do that for them.” In that respect not unlike RPA in back-office functions. “It’s not high value to remember the syntax and the best practice. And even if you have smart compose [in consumer services] it may suggest sentences but can’t read your mind and tell you how you want to respond. So it’s very unlikely to replace you and respond how you would want to. That’s not even in our longest-term plan.”

e.ventures  last year announced a $400 million fund for early-stage investments, and this seems to be coming out of that. With this round, the firm’s general partner Tom Gieselmann is joining the board.

“I’ve been following the developer tools market for over 20 years and believe Codota has distinguished itself as the dominant player in terms of community, product, and technology,” he said in a statement. “We are proud to support Dror and Eran on their mission to transform software development and make coding easier and more efficient for individual developers and teams within the enterprise.”

The post Codota picks up $12M for an AI platform that auto-completes developers’ code appeared first on Artificial Intelligence.

Enterprise artificial intelligence (AI) provider DataRobot has enhanced its platform to include new capabilities in visual AI and automated deep learning, as well as separate capabilities for MLOps and automated time series.

DataRobot’s SVP of product and customer experience, Phil Gurbacki, says the company wants to push the boundaries of what’s possible.

“Subject matter experts from any industry can now solve new business problems by including relevant image-based content along with other more traditional data types. This latest evolution of our platform will empower users to leverage AI to make even better decisions based on broader perspectives.”

According to the company, the new features expand the breadth and availability of AI to more users, simplifying the build and deployment processes for deep learning models.

Visual AI enables users to address computer vision use cases, particularly in areas such as image recognition and classification. A drag-and-drop image process can potentially result in custom deep learning model builds in ‘minutes’, DataRobot says. Further steps can combine images with other data such as dates, raw text, and numeric or categorical data.

A new automated deep learning process enables users to build models ready for production deployment. DataRobot’s deep learning capabilities are powered by a Keras-based model framework.

Furthermore, the platform supports any machine learning model so that the model can be turned into an AI application. An applications gallery also enables business users to find an app that best suits their needs.

DataRobot has also enhanced its MLOps to include drag-and-drop model files developed in languages like Python, so users can deploy them through Kubernetes.

Automated time series features new deep learning techniques to remove forecasting barriers in large-scale multi-series forecasting applications.

Additionally, DataRobot has integrated Paxata’s AI-assisted data preparation solution with its AI catalogue. 

DataRobot acquired Paxata in December 2019 as part of a move to enhance end-to-end AI capabilities.

“From day one, our vision has been to enable enterprises to build a data fabric foundation that helps them achieve their digital and AI transformation,” commented Paxata CEO and cofounder Prakash Nanduri at the time.

“No one platform served the needs of enterprises when it came to finding, prepping, consuming, and governing data and simultaneously building, deploying, and maintaining AI solutions at an enterprise scale – until today. We’ve been working with DataRobot to build the next-generation integrated experience and going forward, our combined entity will have the global scale to deliver on our joint vision.”

The post DataRobot enhances enterprise AI platform appeared first on Artificial Intelligence.

DataRobot, the leader in enterprise AI, announced enhancements to its enterprise AI platform, including AI Applications, Automated Deep Learning, and Visual AI. These new introductions further unlock the value of AI by putting the power of AI into the hands of more users and making it simpler to build and deploy deep learning models.

“Subject matter experts from any industry can now solve new business problems by including relevant image-based content along with other more traditional data types. This latest evolution of our platform will empower users to leverage AI to make even better decisions based on broader perspectives.”

In the latest version of the platform, DataRobot has introduced:

• Visual AI: With Visual AI, users can address computer vision use cases and combine incredibly diverse types of data in their models. Visual AI offers immediate support for use cases requiring image recognition and classification. Users can simply drag and drop a collection of images into a project and build custom deep learning models in minutes. DataRobot’s Visual AI then takes image-based machine learning one step further by allowing users to leverage images alongside any other feature types such as numeric, categorical, dates, and raw text.
• AI Applications: With the latest platform release, any machine learning model, including DataRobot-generated models or models written in R or Python, can be turned into an AI application. This enables employees of all skill levels to interact with the predictive insight of the underlying model and experiment with different scenarios, predict results, and make more informed business decisions. The new feature also includes an Applications Gallery – a one-stop shop that allows business users to find the application that best suits their needs.
• Automated Deep Learning: DataRobot has significantly boosted its deep learning capabilities, powered by a new Keras-based model framework for which DataRobot recently secured a provisional patent. Traditionally, training deep learning models is expensive and time consuming. DataRobot’s new capabilities allow users to build successful and reliable deep learning models that are ready to deploy into production. The new capabilities also make it easy to understand these models – all with the infrastructure a user has in place.

“Having pioneered the automated machine learning category, we are proud to push the boundaries of what’s possible with the technology by offering these novel automated deep learning and Visual AI capabilities,” said Phil Gurbacki, SVP of Product and Customer Experience, DataRobot. “Subject matter experts from any industry can now solve new business problems by including relevant image-based content along with other more traditional data types. This latest evolution of our platform will empower users to leverage AI to make even better decisions based on broader perspectives.”

Additionally, DataRobot has unveiled enhancements to:

• MLOps: In this release, DataRobot MLOps has been enhanced to include pre-packaged model environments so users can drag-and-drop model files, developed in languages such as Python and R, and deploy them using Kubernetes. The release also includes unlimited batch scoring with integrations to leading cloud storage options for massive scale. Lastly, the enhanced MLOps solution offers Monitoring Agents that can capture metrics from models deployed to almost any environment.
• Automated Time Series: Automated Time Series now features new deep learning techniques that remove the traditional forecasting barriers to make easy work of large-scale multi-series forecasting applications.
• DataRobot Paxata: Following the acquisition of Paxata in December 2019, DataRobot has integrated Paxata’s AI-assisted data preparation solution seamlessly with its AI Catalog to empower novice and expert users to rapidly explore, clean, combine, and shape data for training and deploying machine learning models.

The post DataRobot Unveils Latest Version of Enterprise AI Platform, Introducing Visual AI, AI Applications, and Automated Deep Learning appeared first on Artificial Intelligence.

In a recent blog post, Google announced the beta of Cloud AI Platform Pipelines, which provides users with a way to deploy robust, repeatable machine learning pipelines along with monitoring, auditing, version tracking, and reproducibility. 

With Cloud AI Pipelines, Google can help organizations adopt the practice of Machine Learning Operations, also known as MLOps – a term for applying DevOps practices to help users automate, manage, and audit ML workflows. Typically, these practices involve data preparation and analysis, training, evaluation, deployment, and more. 

Google product manager Anusha Ramesh and staff developer advocate Amy Unruh wrote in the blog post: 

When you’re just prototyping a machine learning (ML) model in a notebook, it can seem fairly straightforward. But when you need to start paying attention to the other pieces required to make an ML workflow sustainable and scalable, things become more complex.

Moreover, when complexity grows, building a repeatable and auditable process becomes more laborious.

Cloud AI Platform Pipelines – which runs on a Google Kubernetes Engine (GKE) Cluster and is accessible via the Cloud AI Platform dashboard – has two major parts: 

• The infrastructure for deploying and running structured AI workflows integrated with GCP services such as BigQuery, Dataflow, AI Platform Training and Serving, Cloud Functions, and
• The pipeline tools for building, debugging and sharing pipelines and components.

With the Cloud AI Platform Pipelines users can specify a pipeline using either the Kubeflow Pipelines (KFP) software development kit (SDK) or by customizing the TensorFlow Extended (TFX) Pipeline template with the TFX SDK. The latter currently consists of libraries, components, and some binaries and it is up to the developer to pick the right level of abstraction for the task at hand. Furthermore, TFX SDK includes a library ML Metadata (MLMD) for recording and retrieving metadata associated with the workflows; this library can also run independently. 

Google recommends using KPF SDK for fully custom pipelines or pipelines that use prebuilt KFP components, and TFX SDK and its templates for E2E ML Pipelines based on TensorFlow. Note that over time, Google stated in the blog post that these two SDK experiences would merge. The SDK, in the end, will compile the pipeline and submit it to the Pipelines REST API; the AI Pipelines REST API server stores and schedules the pipeline for execution.

An open-source container-native workflow engine for orchestrating parallel jobs on Kubernetes called Argo runs the pipelines, which includes additional microservices to record metadata, handle components IO, and schedule pipeline runs. The Argo workflow engine executes each pipeline on individual isolated pods in a GKE cluster – allowing each pipeline component to leverage Google Cloud services such as Dataflow, AI Platform Training and Prediction, BigQuery, and others. Furthermore, pipelines can contain steps that perform sizeable GPU and TPU computation in the cluster, directly leveraging features like autoscaling and node auto-provisioning.

AI Platform Pipeline runs include automatic metadata tracking using the MLMD – and logs the artifacts used in each pipeline step, pipeline parameters, and the linkage across the input/output artifacts, as well as the pipeline steps that created and consumed them.

With Cloud AI Platform Pipelines, according to the blog post customers will get:

• Push-button installation via the Google Cloud Console
• Enterprise features for running ML workloads, including pipeline versioning, automatic metadata tracking of artifacts and executions, Cloud Logging, visualization tools, and more 
• Seamless integration with Google Cloud managed services like BigQuery, Dataflow, AI Platform Training and Serving, Cloud Functions, and many others 
• Many prebuilt pipeline components (pipeline steps) for ML workflows, with easy construction of your own custom components

The support for Kubeflow will allow a straightforward migration to other cloud platforms, as a respondent on a Hacker News thread on Google AI Cloud Pipeline stated:

Cloud AI Platform Pipelines appear to use Kubeflow Pipelines on the backend, which is open-source and runs on Kubernetes. The Kubeflow team has invested a lot of time on making it simple to deploy across a variety of public clouds, such as AWS, and Azure. If Google were to kill it, you could easily run it on any other hosted Kubernetes service.

The release of AI Cloud Pipelines shows Google’s further expansion of Machine Learning as a Service (MLaaS) portfolio – consisting of several other ML centric services such as Cloud AutoML, Kubeflow and AI Platform Prediction. The expansion is necessary to allow Google to further capitalize on the growing demand for ML-based cloud services in a market which analysts expect to reach USD 8.48 billion by 2025, and to compete with other large public cloud vendors such as Amazon offering similar services like SageMaker and Microsoft with Azure Machine Learning.

Currently, Google plans to add more features for AI Cloud Pipelines. These features are:

• Easy cluster upgrades 
• More templates for authoring ML workflows
• More straightforward UI-based setup of off-cluster storage of backend data
• Workload identity, to support transparent access to GCP services, and 
• Multi-user isolation – allowing each person accessing the Pipelines cluster to control who can access their pipelines and other resources.

Lastly, more information on Google’s Cloud AI Pipeline is available in the getting started documentation.

The post Google Announces Cloud AI Platform Pipelines to Simplify Machine Learning Development appeared first on Artificial Intelligence.

DUBAI – DataRobot, the leader in enterprise AI, and InterSystems, a global leader in information technology platforms for health, business, and government applications, today announced a partnership designed to accelerate the application of AI in healthcare. Through an integration and reseller agreement, the partnership makes it easier for InterSystems customers to integrate predictions and insights from DataRobot’s enterprise AI platform into their healthcare applications.

The DataRobot enterprise AI platform provides automation across the entire AI lifecycle, accelerating and streamlining a user’s journey from data to value. Through this partnership, the enterprise AI platform will be integrated with InterSystems IRIS® data platform and InterSystems IRIS for Health™, the world’s first and only data platform specifically engineered to extract value from healthcare data.

“Healthcare is prime for disruption in ways that benefit patients, providers and payers, and AI represents the next frontier,” said Bill Hobbib, SVP of Marketing, DataRobot. “Our partnership with InterSystems makes it easier for users to leverage the power of AI to deliver high-quality care, improve patient experience and outcomes, all while reducing the cost of care through AI-driven efficiencies.”

DataRobot’s enterprise AI platform will augment InterSystems IntegratedML, which gives developers access to AutoML capabilities directly from SQL. This allows InterSystems IRIS and IRIS for Health customers to embed predictions within their existing applications in a simple, intuitive, and scalable way. InterSystems’ technology simplifies the data acquisition, normalization, and other “data wrangling,” and DataRobot’s technology simplifies the entire AI process from feature engineering and modeling to deployment and monitoring. This will accelerate delivery and time to value of ML/AI capabilities in real-time decisioning applications.

“DataRobot’s enterprise AI platform is a natural extension across all InterSystems deployments using our IntegratedML, both in healthcare and also in other industries,” said Scott Gnau, head of Data Platforms for InterSystems. “This enables developers to easily include machine learning and AI extensions, creating real-time enhanced decisions and analytics for their applications. It makes operations robust as well. IntegratedML is built into InterSystems IRIS and IRIS for Health – a secure and reliable platform that runs many of the world’s most health-critical applications, and DataRobot brings in best-in-class MLOps.”

About DataRobot

DataRobot is the leader in enterprise AI, delivering trusted AI technology and ROI enablement services to global enterprises competing in today’s Intelligence Revolution. DataRobot’s enterprise AI platform democratizes data science with end-to-end automation for building, deploying, and managing machine learning models. This platform maximizes business value by delivering AI at scale and continuously optimizing performance over time. The company’s proven combination of cutting-edge software and

world-class AI implementation, training, and support services, empowers any organization – regardless of size, industry, or resources – to drive better business outcomes with AI.

With a singular focus on AI since its inception, DataRobot has a proven track record of delivering AI with ROI. DataRobot has offices across the globe and $431 million in funding from top-tier firms, including New Enterprise Associates, Sapphire Ventures, Meritech, and DFJ Growth.

About InterSystems

InterSystems is the information engine that powers some of the world’s most important applications. In healthcare, business, government, and other sectors where lives and livelihoods are at stake, InterSystems has been a strategic technology provider since 1978. InterSystems is a privately held company headquartered in Cambridge, Massachusetts (USA), with offices worldwide, and its software products are used daily by millions of people in more than 80 countries.

The post DataRobot and InterSystems partner to accelerate adoption of ai in healthcare appeared first on Artificial Intelligence.

Run:AI this week announced the general availability of a namesake platform based on Kubernetes that enables IT teams to virtualize graphical processor unit (GPU) resources.

Company CEO Omri Geller says the goal is to enable IT teams to maximize investments in expensive GPUs by leveraging a single line of code to plug in its platform on top of Kubernetes. That would enable IT teams to take advantage of container orchestration to schedule artificial intelligence (AI) workloads across multiple GPUs, and allows certain AI workloads to be prioritized over others, he says.

Geller notes that GPUs don’t lend themselves well to traditional virtual machines. Kubernetes provides an alternative approach to virtualizing bare-metal GPU resources, which are among the most expensive IT infrastructure resource any IT organization can invoke in the cloud or deploy in on-premises IT environments.

Containers are widely employed by AI models because they provide a means to efficiently manage access to large amounts of data without having to update the entire AI application every time an element of that AI model changes. One of the biggest challenges organizations now face is maximizing GPU resources at a time when the number of AI models being built and deployed continues to expand rapidly. Most organizations not only have to frequently update AI models when new data sources become available, but they also discover that over time many AI models need to be replaced altogether, as business conditions and circumstances evolve.

Of course, as those AI models are updated they also need to be slipstreamed into applications, which in turn creates a series of new DevOps challenges for organizations.

For the most part, GPUs today are employed to train AI models because they are more efficient than traditional commercial processors. However, most of the inference engines on which AI models are run are deployed on commercial x86 processors from Intel and AMD. Rival NVIDIA, however, is making a case for replacing x86 processors with GPUs that are becoming less expensive. The Run:AI platform can be employed to maximize GPUs to either train AI models or run inference engines, Geller says.

It’s not at all clear to what degree AI will be injected into applications. There’s no doubt AI will play a significant role in the future of application development. However, most organizations won’t be able to afford to inject AI instantly into every application or re-engineer business processes overnight, so it may still be a while before AI is all-pervasive. In fact, the truth is there is a lot of trial and error involving AI than most data scientists would care to admit.

In the meantime, however, it’s becoming clear Kubernetes will emerge as a de facto platform for both building and deploying AI applications, especially as platforms such as Kubeflow, a set of tools for building AI models, continue to mature. One major issue now is determining how to efficiently manage all the AI workloads that now are heading in the direction of Kubernetes.

The post Run:AI Leverages Kubernetes to Virtualize GPUs appeared first on Artificial Intelligence.

Intel is ending work on its Nervana neural network processors (NNP) in favor of an artificial intelligence line it gained in the recent $2 billion acquisition of Habana Labs.

Intel acquired Nervana in 2016 and issued its first NNP chip one year later. After the $408 million acquisition by Intel, Nervana co-founder Naveen Rao was placed in charge of the AI platforms group, which is part of Intel’s data platforms group. The Nervana chips were meant to compete with Nvidia GPUs in the AI inference training space, and Facebook worked with Intel “in close collaboration, sharing its technical insights,” according to former Intel CEO Brian Krzanich.

For now, Intel has ended development of its Nervana NNP-T training chips and will deliver on current customer commitments for its Nervana NNP-I inference chips; Intel will move forward with Habana Labs’ Gaudi and Goya processors in their place.

There are two parts to neural networks: training, where the computer learns a process, such as image recognition; and inference, where the system puts what it was trained to do to work. Training is far more compute-intensive than inference, and it’s where Nvidia has excelled.

Intel said the decision was made after input from customers, and that this decision is part of strategic updates to its data-center AI acceleration roadmap. “We will leverage our combined AI talent and technology to build leadership AI products,” the company said in a statement to me.

“The Habana product line offers the strong, strategic advantage of a unified, highly-programmable architecture for both inference and training. By moving to a single hardware architecture and software stack for data-center AI acceleration, our engineering teams can join forces and focus on delivering more innovation, faster to our customers,” Intel said.

This outcome from the Habana acquisition wasn’t entirely unexpected. “We had thought that they might keep one for training and one for inference. However, Habana’s execution has been much better and the architecture scales better. And, Intel still gained the IP and expertise of both companies,” said Jim McGregor, president of Tirias Research.

The good news is that whatever developers created for Nervana won’t have to be thrown out. “The frameworks work on either architecture,” McGregor said. “While there will be some loss going from one architecture to another, there is still value in the learning, and I’m sure Intel will work with customers to help them with the migration.”

This is the second AI/machine learning effort Intel has shut down, the first being Xeon Phi. Xeon Phi itself was a bit of a problem child, dating back to Intel’s failed Larrabee experiment to build a GPU based on x86 instructions. Larrabee never made it out of the gate, while Xeon Phi lasted a few generations as a co-processor but was ultimately axed in August 2018.

Intel still has a lot of products targeting various AI: Mobileye, Movidius, Agilex FPGA, and its upcoming Xe architecture. Habana Labs has been shipping its Goya Inference Processor since late 2018, and samples of its Gaudi AI Training Processor were sent to select customers in the second half of 2019.

The post Intel drops work on one of its AI-chip lines in favor of an other appeared first on Artificial Intelligence.

DataRobot has acquired Paxata, a provider of a data fabric and data prepping tools, as part of an effort to create an end-to-end platform for artificial intelligence (AI) applications that spans everything from how raw data is collected to the way an AI application is built and deployed.

While DataRobot plans to continue to make the data prep tools created by Paxata in a stand-alone fashion, the acquisition of the company fills a significant data preparation gap in DataRobot’s ambitions to build an AI platform, says Phil Gurbacki, senior vice president for product management and customer experience at DataRobot.

The first realization of that effort is a new AI Catalog from DataRobot, which makes it easier for business analysts and citizen data scientists to prepare data for machine learning. In fact, the primary reason DataRobot chose to acquire Paxata is that the data prepping tools the company developed for business analysts also adapt themselves to the unique requirements of prepping data for AI applications, says Gurbacki.

In fact, Gurbacki notes any organization building an AI application today will encounter significant data management challenges.

“It’s a critical need for all of our customers,” says Gurbacki.

Unlike in traditional business intelligence (BI) application environments, the need to prep data in AI environments becomes a continuous process. AI models need to be regularly updated and retrained as new data sources become available. To facilitate that process DataRobot plans to leverage the expertise of both companies to define a set of best practices for building pipelines and managing AI applications, says Gurbacki.

That DataRobot platform, adds Gurbacki, not only spans multiple computing frameworks such as Apache Spark, it’s also designed from the ground up to address the needs of multiple classes of end-users, analysts, and developers participating in the AI development processes. The next major initiative will also serve to make the DataRobot platform more programmatically addressable for developers, notes Gurbacki.

As a rule, Gurbacki says DataRobot is fairly selective about the customers it is willing to engage. Far too many organizations appear to want to launch AI projects simply so they can say they have one. DataRobot will vet customer AI project based on the achievability of a clear business outcome before deciding to make any commitment, says Gurbacki.

Paxata is the third acquisition DataRobot has made this year as part of its quest to build an AI platform. Previously, DataRobot acquired ParallelM, which has been integrated into the company’s machine learning operations (MLOps) platform, and Cursor, which has been integrated into the AI Catalog the company just launched. DataRobot thus far has raised $431 million in funding, while Paxata prior to being acquired has raised $90 million.

It’s still too early to say how the AI platform wars will play out. Organizations will need to carefully weigh not just how robust the AI applications they can build are using any platform, but also how quickly those AI applications can be built, deployed, and updated. After all, the useful life of any AI model is likely to be fleeting as not just new sources of data become available, but also the goals of the business continue to evolve and change.

The post DataRobot Acquires Paxata to Extend AI Platform appeared first on Artificial Intelligence.