Helping Developers Make AI Real

The pace of AI developer community growth and advancement is incredible. Kaggle added more than 600,000 new users in 2017, twice as many as in 2016[1]. The Scopus database now contains more than 200,000 papers indexed with the key term “artificial intelligence.”[2] Dozens of new AI submissions are posted to arXiv.org each week. Driven by their innate curiosity, thousands of talented, dedicated developers and researchers are creating the open source tools, frameworks, educational resources, and collaborative communities that are delivering a new era of AI capabilities.

What we build are tools to support you, the developer.  With an eye to this community’s critical role in driving AI technologies forward, I am excited to share more about our first Intel® AI DevCon 2018 (AIDC), May 23rd and 24th at The Palace of Fine Arts in San Francisco.

Why Intel® AI DevCon

At Intel, we’re encouraged by the impact that AI is having, driven by its rich community of developers. AI is mapping the brain in real time, discovering ancient lost cities, identifying resources for lunar exploration, helping to protect Earth’s oceans, and fighting fraud that costs the world billions of dollars per year, to name just a few projects. It is our privilege to support this community as it delivers world-changing AI across verticals, use cases, and geographies.

To drive AI forward, developers want purpose-built technologies, tools, and resources. We know it’s going to take more to deliver the promise AI holds and do the amazing things you hope to achieve.

AIDC will help you make your next move in AI using cost-effective, reliable, versatile Intel® architecture, and tapping the expertise of leaders in data science, machine and deep learning, application development, and AI research. It will inspire you with real-world demos, share the latest breakthroughs, enhance your skills, and better connect you with industry peers. You don’t want to miss it.

What We’re Talking About at AIDC

At AIDC, we’ll go deep in several topics.

Developer Tools

AIDC is the best place to learn about the latest open source software tools to make your AI algorithms performant and portable across different hardware types. For example, Intel® Math Kernel Library for Deep Neural Networks increases the performance of deep neural networks via an open source library. Optimizations for running TensorFlow* and MXNet* on Intel Xeon processors have delivered radical improvements in time to solution for both training and inference. Intel® nGraph™, an open source compiler sitting between the framework and hardware layers, helps to future-proof work so developers have the flexibility to use new frameworks, models, and hardware as they arise.

Hardware

At AIDC, you will learn how Intel® Xeon® Scalable processors are enabling users to achieve more with AI and data center workloads running side by side. Additionally, we will be highlighting hardware solutions featuring Intel® FPGAs for flexible, real-time, low-latency inference.  We are also excited to share updates and progress on our first Intel® Nervana™ Neural Network Processor family.  But that’s not all, join us to discover the possibilities for using Intel’s wider portfolio: low-power processors, from Intel® Movidius™ for AI at the edge, the Intel®-based Amazon Web Services* (AWS*) DeepLens* for deep-learning wireless video applications, and more.

Community

It’s inspiring how quickly developments move from research to the source code of an AI model. We’re seeking more ways we can help support the vibrant knowledge-sharing community around AI. The team from Intel AI Labs will be sharing the development around open and flexible libraries for Reinforcement Learning, pruning and quantization, and Natural Language Processing at different abstraction layers, all with the goal of making developer’s lives easier. We want to hear from you, too, and get feedback on our progress in supporting this community.

Key Activities at AIDC

Here are a few of the AIDC sessions and activities I’m most excited about:

Sessions:

  • Designing Far-Field Speech Processing Systems with Intel and Alexa* Voice Service with Fletch Nelson, Amazon – Learn how to integrate Alexa capabilities directly into your product via a far-field voice-initiated prototype with the Alexa Voice Service (AVS), and walk out with an Alexa-enabled development kit
  • AWS DeepLens Workshop: Building Computer Vision Applications with Nagajyothi Nookula and Mike Miller (Amazon Web Services) – A hands-on session helping developers of all skill levels build and deploy computer vision models using AWS DeepLens
  • TensorFlow Optimized for Intel Xeon Processors with Vivek Rane, Intel – An overview of Intel’s optimizations for TensorFlow on Intel Xeon processors, including specific tuning tips for developers to deliver the best performance
  • FPGAs in Deep Learning Inference Today with Adrian Macias, Intel – A look into customer success stories for FPGA deployment in deep learning, including discussion on the hardware and software stack abstractions that are delivering FPGA value for deep learning
  • Build a DIY smart security camera using the Intel Movidius Neural Compute Stick with Ashwin Vijayakumar, Intel – A practical session describing how to build a DIY smart security camera using an Intel Movidius Neural Compute Stick, a Raspberry Pi, and a deep neural network based object detector

Demos:

  • Doctor Hazel – An AI that can find melanoma by looking at a patient’s skin, powered by an Intel Movidius Neural Compute Stick
  • Clean Water AI – A potentially life-saving applications that detects when water has been contaminated by harmful bacteria, also based on an Intel Movidius Neural Compute Stick
  • Ziva Dynamics Interactive Characters – Interact with the world’s most advanced out-of-the-box virtual characters and see how they’re being used in cutting-edge film, game, and VR experiences

Join us at AIDC

We’re hard at work and have a lot of great things to share with you at AIDC. We’ll look forward to seeing you there so you can learn how to make your next move in AI on Intel architecture.

See you in San Francisco!

Notices and Disclaimers

cIntel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration. No computer system can be absolutely secure. Check with your system manufacturer or retailer or learn more at intel.com.

Intel, the Intel logo, Xeon, Movidius, and nGraph are trademarks of Intel Corporation or its subsidiaries in the U.S. and/or other countries.

*Other names and brands may be claimed as the property of others.

© Intel Corporation 

[1] http://blog.kaggle.com/2017/12/26/your-year-on-kaggle-most-memorable-community-stats-from-2017/

[2] https://www.forbes.com/sites/louiscolumbus/2018/01/12/10-charts-that-will-change-your-perspective-on-artificial-intelligences-growth/#1f6722014758