When choosing pipeline components, try to build a pipeline where each component is a relatively simple function that can be learned from only a modest amount of data.

Google

Task simplicity can favor adding an intermediate component instead of feeding raw inputs into a later component. In the self-driving example, adding a lane-marking detector gives the planner missing information while avoiding a path-planning module that must solve a more complex raw-image task.

Task Simplicity for Pipeline Component Selection

Machine Learning Yearning, a free ebook from Andrew Ng, teaches you how to structure Machine Learning projects.
https://www.deeplearning.ai/machine-learning-yearning/

Note: The content of the book is aligned with the Coursera Deeplearning.ai specialization.  https://www.deeplearning.ai/deep-learning-specialization/ 

Machine Learning Yearning (Deeplearning.ai)

Even among binary image classification tasks, some tasks are easier for a neural network to learn than others. Easier tasks can be learned with fewer training examples.

Easier Component Tasks Need Fewer Training Examples

Breaking a complex task into simpler subtasks and explicitly coding the subtask steps gives the algorithm prior knowledge. That prior knowledge can help it learn the task more efficiently.

Learn Before

Related