Monday, December 4, 2017

Article: Has Deep Learning Made Traditional Machine Learning Irrelevant? by William Vorhies

To go to the article, click here.

The article makes statements like
  1. ... CNNs and RNNs are very difficult to train and sometimes fail to train at all ...
    1. ... If you are building a CNN or RNN from scratch you are talking weeks or even months of development time ...
      1. ... CNNs and RNNs require extremely large amounts of labeled data on which to train which many companies find difficult or too costly to acquire ...
        1. ... two distinct data science markets that have evolved ... ‘Big Web User’ ... But upwards of 80% of the application of data science today is still in the prediction of consumer behavior ...
          1. ... need to see these competitions [Kaggle] like Formula One racing ...
          Long winded way of saying approach neural networks with caution.