The First Rule of Machine Learning: Start without Machine Learning

Share This Post

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

More To Explore


An Eerie Poisonous Fog

If you’re visiting a farm, keep an eye out for an ominous-looking bronze fog. Nitrogen tetroxide, a red-orange gas, is made by certain bacteria during fermentation when oxygen is lacking in the crop. The lack of oxygen causes the bacteria to use the nitrate within the crops, which then produces nitrogen tetroxide. Inhalation of the gas is known to cause permanent lung damage or kill people and livestock. To prevent this, farmers have started performing nitrate tests after four weeks of fermentation. Read full article here


Happiness Is Two Scales

It’s put forth that happiness and unhappiness should be viewed as two different ‘scales’, not either end of one scale. For instance, if you’re not satisfied with your life, try to identify the root problem: do you not have enough happiness, or are you just unhappy? Depending on your answer to the question, the solutions may differ. Read full article here

Do You want to embrace intellectual freedom and join our premium users?


The occasional email full of conversation-worthy content