Deep Learning is a category of machine learning with special advantages for some tasks and disadvantages for others.
Machine learning workflows begin by identifying features within data sets. For structured information with relatively few columns and rows, this is straightforward. Most practical business predictions such as classification and regression fall into this category.
Unstructured data, such as image and voice, have vast numbers of “features” in the form of individual pixels or wave forms. Identifying those features to structured AI algorithms is tedious or impossible. Deep Learning is a technique where the AI algorithm itself extracts progressively higher levels of feature recognition, passing information through potentially hundreds of neural network layers. Deep learning algorithms power image and speech recognition for driverless cars and hand-free speakers
Plusses of Deep Learning
- Scale – Deep learners can handle vast amounts of data, and they always improve with more data. Shallow learning converges and stops improving with additional data.
- Dimensions – Deep learners can move past the limitations of a few hundred columns to perform well on very wide structured data set.
- Non-Numeric – Deep learning brings AI into the human realm of speech and vision, which serve people in new and valuable ways.
Minuses of Deep Learning
- Training Data – Deep learners need labeled data from which to learn. Amassing sufficient examples for recognition accuracy to be learned can be daunting.
- Not for Small Data Sets – Data sets that are too simple or too small cause deep learners to fail by overfitting.
- Resource Consumption – Deep learning on vast data stores can require days or weeks of processing on a single problem.
The take-away: Deep learners are great for unstructured data and may be useful for classification with large and detailed structured data sets. Squark includes deep learners in the stack of algorithms it uses for AutoML. You will know from the Squark Leaderboard whether deep learning was a winner.