In recent times, we take sources similar to Wikipedia or Reddit with no consideration — these sources depend on the collective information of particular person contributors to serve us with principally correct info, which is typically known as the “knowledge of the gang”. The concept is that the collective choice might be extra correct than any particular person’s judgement, since we will every have our personal implicit biases and/or lack of awareness, leading to some stage of error in our judgement. Collectively, these errors is perhaps offset by one another — for instance, we will compensate for another person’s lack of awareness/experience in a single space, whereas they’d make up for ours in different areas. Making use of this concept to machine studying leads to “ensemble” strategies.
At a really excessive stage, we prepare machine studying fashions with a purpose to make predictions concerning the future. In different phrases, we offer the fashions with the coaching information with the hope that mannequin could make good predictions concerning the future. However what if we may prepare a number of machine studying fashions after which one way or the other combination their opinions concerning the predictions? It seems, this could be a very helpful method, which is broadly used within the business.