Why Simple Machine Learning Models Are Key To Driving Business Decisions. Posted 21_10_17
What Is It? Tiny Model Theory is the idea that you don’t need to use heavy-duty ML models to carry out simple repetitive everyday business predictions.
Why Important: Using more lightweight models cuts down on the time you’d need for the aforementioned bottlenecks, decreasing your time to value. Tiny models, like logistic regression, can train concurrently by making use of distributed ML that parallel trains models across different cloud servers. Tiny are ideal candidates for distributed ML. Some of the top companies prefer simple models for their distributed ML pipeline involving edge devices, like IOTs and smartphones.