Thank you to all the Intel® AI Builders and event attendees who joined us at the O’Reilly Artificial Intelligence Conference in New York City to make this anniversary showcase for the Intel® AI Builders program such an immense success. The event featured rapid-fire demonstrations and presentations that showcased Intel® AI Builders solutions in action across financial services, retail, and healthcare sectors, along with cross-industry AI solutions. The event also served as a powerful networking opportunity for both attendees and showcase presenters.
Download the full presentation here
Recordings of the individual presentations given by the following Intel® AI Builders members can be found below:
We are excited to have received incredibly positive feedback and are happy that this event was truly valuable to both Intel AI Builders presenters and event attendees.
“Best event I’ve been to in years of going to O’Reilly events.” - Intel® AI Builders Showcase attendee
“As a result of our participation in the Intel® AI Builder Showcase at the O'Reilly AI Conference in New York, we had numerous valuable leads that resulted in several strategic meetings. I want to thank you for your assistance in helping us prepare in advance as well. It was terrific and look forward to our continued collaboration.” - Josh Ledeenavp, Hi-Tech Business Unit, Mphasis
“Very good for networking ... we should do a business case any event that the Intel® AI Builders team is doing - we want to be wherever they are.” - Dan Klein, Chief Data Officer, Valtech
“A barometer of what’s happening in the AI ecosystem.” - Intel® AI Builders Showcase attendee
Featured Intel® AI Builders Presentations
H2O.ai is an open source leader in AI with a mission to democratize AI. H2O.ai is transforming the use of AI with software with its category-creating visionary open source machine learning platform, H2O. More than 18,000 companies use open-source H2O in mission-critical use cases for finance, insurance, healthcare, retail, telco, sales, and marketing. H2O.ai’s Driverless AI on Intel® Xeon® processors is an automatic machine learning platform that uses AI to provide an easier, faster and more effective means of implementing data science. H2O Driverless AI employs the techniques of expert data scientists in an easy-to-use application that helps scale your data science efforts. We will showcase its capabilities by predicting credit card payment default using a Taiwanese bank's dataset from a Kaggle experiment.
When Intel® AI Builders partner Mphasis set out to build Mphasis DeepInsights™, a Cognitive Intelligence platform for enterprise customers, they chose Intel to help them reach their goals. Mphasis DeepInsights™ is a patent pending Cognitive Intelligence platform, which enables enterprises to have faster and more effective access to insights from data. Cognitive intelligence is one of the major pillars on which tomorrow's enterprises are being built. DeepInsights™ is powered by state-of-the-art algorithms in machine learning, neural network, deep learning, semantics and more. Mphasis and Intel have worked together to optimize DeepInsights for Intel architecture and to ensure that infrastructure solutions are as simple and interoperable as possible. The DeepInsights cognitive platform runs in a Microsoft Windows environment and relies on Intel® Distribution for Python and Intel® Optimization for TensorFlow. This further enables enterprises to engage with their customers through personalized experiences and explore newer business models.
The faster and smarter we process and analyze data, the more relevant and meaningful the insights and the higher the profits. Just having data is no longer useful and traditional ways of thinking about analytics no longer apply. To drive business impact, it is important to analyze your time-sensitive data in the moment for efficient operations, regulatory compliance, and enhanced customer experience. In fact, real-time data processing and advanced analytics are becoming a necessity. GigaSpaces InsightEdge, an in-memory real-time analytics software platform, is powering faster transactional and analytical processing. Gigaspaces will showcase how the software platform leverages Intel’s Analytics Zoo and Intel® Optane™ DC Persistent Memory to operationalize real-time ML and DL for real-time stock price prediction based on sentiment analysis.
The team from Altoros will share its experience with developing machine learning–based solutions that automate workflows within insurance, car rental/maintenance, finance, publishing, and other industries. The team will showcase two solutions developed using Intel® Xeon® Scalable processors and the Intel® Distribution of OpenVINO™ toolkit on the Google Cloud Platform. One solution was developed to facilitate the process of assessing car damage right at the accident scene; the other helps to automate recognition, extraction, and analysis of meaningful information from PDF-based documents. Attendees will learn real-life implementation scenarios and how both solutions can be integrated into the existing workflows of insurance, car rental/maintenance services, finance, publishing, and other industries.
DataRobot, powered by Intel® Xeon® Scalable processors, is the first automated machine learning solution that enables enterprises to become AI-driven and build machine learning models at the click of a button. DataRobot automates machine learning to deliver a productivity boost to organizations of all sizes in order to maximize value from their data analytics endeavors. Intel CPUs bring power and scalability, and the ability to build production-scale machine learning models on AI-optimized infrastructure. We will examine the automated ML platform within the context of a personal loan default case and see how DataRobot automates the key tasks while applying best practices as well as guardrails throughout a typical machine learning process. We will also look at various insights and complementary functionalities towards model and prediction results explainability. Listen to DataRobot’s Intel on AI podcast episode to learn more.
Supervised machine learning requires large labeled datasets - a prohibitive limitation in many real-world applications. What if machines could learn with fewer labeled examples? At Cloudera Fast Forward Labs we translate machine learning breakthroughs in academia for industry applications. This talk explores and demonstrates an algorithmic solution that relies on collaboration between humans and machines to label smartly and discuss product possibilities for various verticals.
Valtech will showcase a live demonstration of a prototype that demonstrates how AI is able to recognize human behavior in retail stores. The solution for integrated location, face, and apparel tracking in real-time is able to tie customer data (location, face metadata, and apparel metadata) in the floor area around a point-of-sale or digital signage, allowing for behavioral insights, experimentation, and validation of customer engagement. The initial prototypes used Intel® Movidius™ to enable deep learning across video streams. With Intel’s assistance, we subsequently joined our prototypes together to enable store location tracking, clothing detection and gender/sentiment assessment. These separate prototypes were then ported into a single pipeline using the Intel® Distribution of OpenVINO™ toolkit.
AAEON is focused utilizing AI on the edge to resolve challenges including latency, network bandwidth, reliability and security, which arose when developers tried to bring all AI tasks to the cloud. UP AI Edge offers a solution to cloud limitations by bringing AI performance and hardware acceleration not ‘at’ but ‘ON’ the Edge of the Internet of Things and the latest technologies. It includes modules powered by multiple Intel® Movidius™ Myriad™ X VPUs and certified systems with Intel® Movidius™ Myriad™ X inside. A complete eco-system, from connectivity and software tools to AI development partners, are at the developers service. Learn about the AAEON on the edge solutions and the path to production.
John Snow Labs:
John Snow Labs' Spark NLP is an industrial-grade Python, Scala, and Java library built on top of Apache Spark* and using Intel® Optimization for TensorFlow*. It provides scalable, trainable, and highly optimized implementations of natural language processing ML and DL pipelines. Learn about recent achievements in productizing NLP Architect by Intel® AI Lab models into Spark NLP which provides the open source community with production-ready versions of some of the most exciting recent research results in natural language understanding, under simple Python & Scala APIs. We will present real-world case studies of applying the library in healthcare, finance, and life science.
AI-based controllers for electro-mechanical systems are proving to be very effective. This requires training of multiple agents and solvers in parallel, at scale. As an artificial intelligence consultancy with world class expertise in deep learning, Minds.ai will present DeepSim, a platform built on Intel® Xeon processors to design and deploy such controllers. We will also show comparisons against results achieved using conventional controller design techniques.
The DeepSim platform can scale from a desktop to a enterprise or public cloud. DeepSim is optimised to run on Intel Xeon servers using Intel's Math Kernel Library. Deep Neural Network primitives provided by MKL greatly helped with scaling and performance optimization.
Medical experts spend significant time and effort in diagnosing medical images. Wipro has developed a medical image segmentation & diagnosis solution using deep learning platform that can assist in the process of image segmentation and diagnosis, which improves speed and productivity. This solution was developed using 8th Generation Intel® Core™ i7 processors and Intel® Distribution of OpenVINO™. It segments the lung area in a CT scan and classifies chest/lung x-rays and CT scans as either “normal” or “abnormal” for diagnostics. We have seen high accuracy with this solution, with the accuracy at about 97% on segmentation and over 90% on classification.
Mobiliya (a QuEST Global Company):
As a global software engineering services (SI) company focusing on next-gen technologies to accelerate digital transformation for the world’s leading companies, we will showcase our capabilities in AI, Vision and Analytics with Intel technologies. For example, in medical imaging, Computer Aided Detection systems are available for detection of tumors and diagnosis of diseases from images acquired by imaging modalities like CT, MRI, X-ray and ultrasound. AI techniques used in computer vision such as object detection and segmentation offer unique possibilities to help radiologists make diagnosis faster and more accurately, leading to prioritization of cases, better patient care, and reduction in cost. When trying to balance accuracy and complexity of the model and real time performance, deployment on Intel® architecture offers an affordable solution. These models can be ported using Intel’s Distribution of OpenVINO toolkit to generate the optimized model to run on Intel processors.
Anaconda Enterprise combines core AI technologies, governance, and cloud-native architecture to enable organizations to automate AI at speed and scale. Learn how to leverage Anaconda Enterprise across the deep learning lifecycle from model building using Intel® Optimization for TensorFlow* to deployment on CPUs without sacrificing speed.
The BlueData EPIC™ (Elastic Private Instant Clusters) software platform solves the infrastructure challenges and limitations that can slow down and stall AI and Big Data deployments. With EPIC software, you can spin up containerized environments for distributed Machine Learning and Analytics – in minutes rather than months. Learn how you can spin up instant ML / DL environments using containers – while ensuring enterprise-grade security and performance. Provide your data science teams with on-demand access to the tools and data they need – whether on-premises or in the cloud. This demo will specifically show how you can deploy containerized ML / DL environments (e.g. for H2O Driverless AI, Intel® Optimization for TensorFlow*, and BigDL with Apache Spark*) using the BlueData EPIC software platform – highlighting the inference aspects of large-scale distributed data pipelines.
Learn about our natural language-based customer service, product guidance, and cross-sell/up-sell solutions. This demo includes a live walkthrough of how general-purpose natural language understanding can be customized for banking using domain-specific AI. The audience will see the end-user experience with the virtual assistant, the creation of the banking specific domain which facilitates transfer learning, and an overview of the virtual assistant creation process. Learn how Avaamo utilizes Intel® Xeon® Scalable processors to execute domain-specific NLU, transfer learning, conversational intelligence, domain modeling, flow modeling, and conversation validation.