Intel on AI

In this Intel on AI podcast episode: Vinod Prasad, General Manager & Lead Architect at Wipro Technologies, joins Intel on AI to discuss Wipro medical imaging solutions powered by Intel Architecture and optimizations. He discusses how the medical industry faces many challenges today including the sometimes difficult and time consuming process of diagnosis of patient chest X-rays. Vinod highlights how Wipro is working to help healthcare providers save time and resources while also improving the patient experience. He describes how Wipro is automating medical image segmentation, analysis, and captioning using deep learning (DL) neural networks running on Intel architecture and optimizations. Utilizing Intel technologies like Intel Distribution of OpenVINO toolkit has enabled Wipro to provide their customers with very efficient and cost effective image analysis solutions on premise and at the edge. To learn more, visit: wipro.com And: builders.intel.com/ai/social-hub/podcast/medical-imaging-wipro-intel Visit Intel AI Builders at: builders.intel.com/ai

16 Mai 201912min

In this Intel on AI podcast episode: FPGA (field-programmable gate array) technology can offer a very high level of flexibility and performance, with low latency. Yet, with software writing thresholds, limited performance optimization, and difficult power control, FPGA solutions can also be challenging to implement. Bob Anderson, General Manager of Sales for Strategic Accounts at Inspur, joins Intel on AI to talk about the Inspur TensorFlow-supported FPGA Compute Acceleration Engine (TF2). Bob illustrates how the TF2 helps customers more easily deploy FPGA solutions and take advantage of the customization and performance of FPGAs for AI inference applications. He also describes how the TF2 is especially suitable for image-based AI applications with high real-time requirements. To learn more, visit: inspursystems.com Visit Intel AI Builders at: builders.intel.com/ai

13 Mai 20198min

In this Intel on AI podcast episode: Leonard Sheiba, the Founder and CEO of Kibernetika, joins Intel on AI to talk about the challenges many organizations face when developing artificial intelligence (AI) solutions especially within an enterprise environment at scale. He points out that AI enabled software components like datasets, models, training, and serving involve continuous validation and retraining to work properly and require maintaining a strict process flow to develop in an enterprise environment. Leonard illustrates how the Kibernetika Machine Teaching Engine is an end-to-end platform that manages the entire AI development workflow from concept through decommission. He describes how their platform can serve many organizations and industry verticals including telecom, healthcare, and nearly any enterprise AI deployment. Lastly, Leonard highlights how the Kibernetika platform runs on Intel Xeon Scalable processors driving the power and efficiency that Kibernetika’s customers need. To learn more, visit: kibernetika.ai Visit Intel AI Builders at: builders.intel.com/ai

10 Mai 20199min

In this Intel on AI podcast episode: As the amount of data grows exponentially businesses are looking to leverage artificial intelligence (AI) to make real-time decisions with it. Yoav Einav, Vice President of Product at GigaSpaces, joins Intel on AI to discuss how the GigaSpaces InsightEdge platform is enabling enterprises to truly take advantage of their data utilizing in-memory computing for fast data analytics and extreme transaction processing. He illustrates how in-memory analytics allows customers to run analytics workloads on real-time data which provides low latency, high-throughput transaction, and co-location of applications and analytics to act on time-critical data in real-time. Yoav explains how enterprises in many different verticals including financial services, retail, transportation, telecom, healthcare, and others can greatly benefit from this capability. Also, Yoav talks about how Intel Optane Data Center persistent memory provides the foundation for the GigaSpaces InsightEdge platform driving more memory capacity per server driving smarter, faster insights based on larger and more varied data sets. To learn more, visit: gigaspaces.com Visit Intel AI Builders at: builders.intel.com/ai

9 Mai 201914min

In this Intel on AI podcast episode: Dr. Gourab De, Vice President of the Data Science at DataRobot, joins Intel on AI to talk about how DataRobot is helping developers and data scientist build AI models without having to write a single line of code. The DataRobot automated machine learning platform delivers the power of artificial intelligence beyond advanced programmers who are able to work through complicated code to a whole new enterprise audience that can greatly benefit from AI. Dr. De highlights that many different industry verticals that are utilizing the DataRobot platform including banking, healthcare, retails, manufacturing and many more. He also discusses how the DataRobot platform utilizes Intel Xeon Scalable processors as a powerful hardware foundation to deliver optimized performance for organizations in pursuit of low latency and scalability – two key performance features to support the AI-driven enterprise. For more information, visit: datarobot.com Visit Intel AI Builders at: builders.intel.com/ai

8 Mai 201913min

In this Intel on AI podcast episode: For this Intel on AI podcast episode we are joined by Vinod Iyengar, Vice President of Data Science Transformations at H2O.ai. Vinod highlights some of the challenges that enterprises face today when implementing AI deployments and how H2O.ai is empowering organizations to overcome those hurdles. He discusses how H2O.ai platforms leverage the performance of Intel Xeon Scalable Processors as well as important optimizations like Intel Math Kernel Library (Intel MKL) and Intel Data Analytics Acceleration Library (Intel DAAL) to bring rapid insights to their customers. Also collaborating on Project Blue Danube and Intel AI Builders, Vinod illustrates how together, Intel and H2O.ai are accelerating AI for everyone and helping to democratize AI. For more information, visit www.h2o.ai/. Visit Intel AI Builders at: builders.intel.com/ai

7 Mai 201916min

In this Intel on AI podcast episode: Ben Vigoda, the founder and CEO of Gamalon, joins the Intel on AI team to talk about how Gamalon is driving the cutting edge of artificial intelligence and machine learning. He highlights their concept of how idea learning or Bayesian Program Synthesis models can enable enterprises to implement incredible understanding of text-based customer communications. Ben describes how Gamalon’s Idea Learning technology learns rapidly and accurately from sparsely labeled training data and enables data scientists to simply enter a series of queries to build a decision tree on the fly based on the text. For more information, visit: gamalon.com Visit Intel AI Builders at: builders.intel.com/ai

14 Mar 201917min

In this Intel on AI podcast episode: Dr. David Ellison, the Senior Artificial Intelligence Data Scientist in the HPC Business Unit at Lenovo joins Intel on AI to discuss the Lenovo Intelligent Computing Orchestration (LiCO). Dr. Ellison explains how LiCO accelerates artificial intelligence (AI) training and traditional high performance computing (HPC) deployment by providing a single software solution that simplifies resource management and the use of a cluster environment. He discusses an exciting real-world deployment where data scientists use LiCO to help analyze satellite images of crop fields to drive drought prevention and increase food security. Another deployment Dr. Ellison highlights is the use of LiCO to help evaluate assembly line product quality implementing computer vision to detect product defects or damage in real time. Dr. Ellison also talks about the great collaboration that Lenovo and Intel are doing through the Intel AI Builders program leveraging resources to help drive software optimizations and improve AI workload performance. To learn more, visit: lenovo.com/ai Visit Intel AI Builders at: builders.intel.com/ai

8 Mar 201912min