Hani Mounla

Don’t make this big machine learning mistake: research vs application


These days, everyone’s getting in on Machine Learning (ML). It’s definitely a great direction to pursue for many businesses since it gives them the ability to deliver tremendous value in a fairly quick and easy way. The demand for machine learning skills is at an all time high. There’s a nice comprehensive report done by McKinsey about how AI is shaping industries and where the opportunities are.

And so we see every business around say:

Hey, we need a machine learning research team really quick! We’ll get the best scientists with lots of publications and pay them lots of money so we can get some machine learning in our business. Hooraaayyy!

But wait just a minute… As a business, do you really need a machine learning research team? Will your business even use them effectively for today’s high price? How much do you even need Machine Learning at all? Is it really that complicated?

If you’re more of a technical person, do you go full throttle on learning how to do machine learning research?

To answer this question we need to differentiate between the two types of ways we can really work with machine learning: research and application.

Machine learning research

Machine learning research is really all about the science. A machine learning researcher is trying to push the boundaries of science, specifically in the field of Artificial Intelligence. These people typically have a Masters or PhD in CS and have many publications in top machine learning conferences. They’re super popular in the research space!

The machine learning researcher is fantastic if you’re doing something really cutting edge. These people are used to finding custom scientific solutions to your problems. If you were to tell them “We’re really good at automatically detecting human intruders using face recognition with 95% accuracy. Could you get us up to 97%?”. The ML researcher is your go to guy!

Machine Learning researchers know this stuff

Here’s the catch: this person likely hasn’t ever actually deployed software into production! They probably aren’t experts at delivering Software as a Service (SaaS) or as a product to your customers, translating the research into practice. They won’t know how to properly package, productionalize, and ship.