Press & Events

How to Apply and Optimize Your Algorithm When You're Ready to Run With AI

Walking With AI: How to Spot, Store and Clean the Data You Need

April 2018

Using specific real-world use cases, Sourav details his process for casting business problems as machine learning problems—from identifying which business problem to solve to hypothesis generation, to AB testing.

In this talk, Sourav discusses a number of case studies with the AI + IoT for Triage business pattern, from digital therapeutics to predictive maintenance. He shares specific examples of the challenges that exist when applying machine learning to IoT sensor data streams and of putting humans in the loop.

One of the most common business patterns of artificial intelligence we see in the field is “AI as Triage”. This is where AI is augmenting human decisions with predictions. The pattern is particularly prevalent when AI is coupled with new data for IoT sensors — whether from medical devices or oil fields.

Sourav details Manifold's 6-step Lean AI process for enterprises to implement AI services to their business. He also discusses how to cast business problems as machine learning problems—from identifying which business problem to solve, to hypothesis generation, to A/B testing.

Using specific real-world use cases, Sourav details his process for casting business problems as machine learning problems—from identifying which business problem to solve to hypothesis generation, to AB testing.