How and When to use Deep Learning

Summary of: “How and When to use Deep Learning” held on Monday, December 17, 2018 from 8 to 10 AM.

In domain after domain, Deep Learning is outperforming both people and competing algorithms at practical tasks:

• ImageNet Hit@5 object recognition error rates have fallen 88% since 2011 and can now recognize objects in photos faster and better than you can
• All major speech recognition engines (Google's, Baidu's, Siri, etc.) automatically translate a speaker's voice in one language to the same voice speaking another language.
• Machines can now beat you at Atari and Go

These breakthroughs are visible as both product offerings (Siri, Real-time Ray-tracing on NVIDIA Turing cards, Netflix Recommendations, etc.) as well as breakthrough results on international open benchmarks. All successful deep learning product teams:

1. Identify the best-matched organizational opportunities to drive ROI with machine learning
2. Address the talent crunch for deep learning expertise
3. Warehouse sufficiently large labeled data sets for training and
4. Update their models at scale as data distributions drift over time

We reviewed all four of these considerations, understood some of the key discriminators between product success and failure, and learned how to identify opportunities best suited for the application of Deep Learning.

‍Dr. John Kaufhold, data scientist and managing partner at Deep Learning Analytics, will lead the discussion. Deep Learning Analytics was one of the four fastest growing companies by revenue in Arlington, Virginia in 2015, and again in 2016. Dr. Kaufhold also serves as Secretary of the Washington Academy of Sciences and is a regular contributor to the DC Data Community, where he moderates the DC2 Deep Learning Discussion list. Prior to founding Deep Learning Analytics, Dr. Kaufhold investigated deep learning algorithms at NIH, SAIC and at GE's Global Research Center. Dr. Kaufhold is named inventor on 10 issued patents in image analysis, and author/coauthor on 40+ publications in the fields of machine learning, image understanding and neuroscience.