I am reading the book Deeplearning by Goodfellow. There it explains about three factors about generalization, which I find it quite blury to imagine.
Excluded the true data generating process—corresponding to underfitting and inducing bias.
Matched the true data generating process.
Included the generating process but also many other possible generating processes—the overfitting regime where variance rather than bias dominates the estimation error.
Please care to explain these three factors. Thank you inadvance!