the no free lunch theorem and the philosophy of deep learning

[beroal]

The no free lunch theorem has made it clear that there is no best machine learning algorithm, and, in particular, no best form of regularization. Instead we must choose a form of regularization that is well suited to the particular task we want to solve. The philosophy of deep learning in general and this book in particular is that a wide range of

Comments

[bluxer]

It definitely should. Quite.

[thedeemon]

That's not what they say here.

They say method M is effective (not best!) on wide range of tasks (not all!).

[beroal]

Unfortunately, “effectively” and “wide range” aren't defined, so either the statement is empty or I may interpret it as I interpreted it.