The First Rule of Machine Learning: Start without Machine Learning

According to various machine learning (ML) practitioners, the first step you must take on the journey to implementing ML is to start not with ML, but with data. Understanding and gathering data is integral to machine learning, and it’s unlikely you’ll have usable data when you’re starting from scratch. How does one solve problems without using ML then? They recommend starting with simple non-ML heuristics; imperfect solutions that work well enough on their own – such as “if/else” rules – in order to test the waters.
Read full article here


The occasional email full of conversation-worthy content