Hands-On Machine Learning for Algorithmic Trading

DL as representation learning

Many ordinary tasks require knowledge about the world. One of the key challenges is to encode this knowledge so a computer can utilize it. For decades, the development of ML systems required considerable domain expertise to transform the raw data (such as image pixels) into an internal representation that a learning algorithm could use to detect or classify patterns.

Similarly, we saw that how much value an ML algorithm adds to a trading strategy depends to a great extent on our ability to engineer features that represent the predictive information in the source data so that the algorithm can process it. Ideally, the representation captures independent drivers of the outcome, as we discussed in Chapter 4, Alpha ...