Edited, memorised or added to reading queue
on 28-Jul-2020 (Tue)
Do you want BuboFlash to help you learning these things? Click here to log in or create user.
| status | not read | reprioritisations | ||
|---|---|---|---|---|
| last reprioritisation on | suggested re-reading day | |||
| started reading on | finished reading on |
| status | not read | reprioritisations | ||
|---|---|---|---|---|
| last reprioritisation on | suggested re-reading day | |||
| started reading on | finished reading on |
| status | not read | reprioritisations | ||
|---|---|---|---|---|
| last reprioritisation on | suggested re-reading day | |||
| started reading on | finished reading on |
Flashcard 5640317701388
| status | not learned | measured difficulty | 37% [default] | last interval [days] | |||
|---|---|---|---|---|---|---|---|
| repetition number in this series | 0 | memorised on | scheduled repetition | ||||
| scheduled repetition interval | last repetition or drill |
| status | not read | reprioritisations | ||
|---|---|---|---|---|
| last reprioritisation on | suggested re-reading day | |||
| started reading on | finished reading on |
How to explain gradient boosting
t boosting Brought to you by explained.ai Terence Parr and Jeremy Howard (We teach in University of San Francisco's MS in Data Science program and have other nefarious projects underway. You mig<span>ht know Terence as the creator of the ANTLR parser generator . Jeremy is a founding researcher at fast.ai , a research institute dedicated to making deep learning more accessible.) Please send comments, suggestions, or fixes to Terence . Contents Roadmap Distance to target An introduction to additive modeling An introdu
then we have 
and can abstract away the individual terms, also as functions, giving us the addition of three subfu
| status | not read | reprioritisations | ||
|---|---|---|---|---|
| last reprioritisation on | suggested re-reading day | |||
| started reading on | finished reading on |
Gradient boosting: Distance to target
, which leads us to the final plot matching our target function: Decomposing a complicated function into simpler subfunctions is nothing more than the divide and conquer strategy that we program<span>mers use all the time. In this case, we are dividing a potentially very complicated function into smaller, more manageable bits. For example, let's call our target function then we have and can abstract away the individual terms, also as functions, giving us the addition of three subfunctions: where: More generally, mathematicians describe the decomposition of a function into the addition of M subfunctions like this: The sigma notation is a for-loop that iterates m fr
| status | not read | reprioritisations | ||
|---|---|---|---|---|
| last reprioritisation on | suggested re-reading day | |||
| started reading on | finished reading on |
Gradient boosting: Distance to target
ore manageable bits. For example, let's call our target function then we have and can abstract away the individual terms, also as functions, giving us the addition of three subfunctions: where: <span>More generally, mathematicians describe the decomposition of a function into the addition of M subfunctions like this: The sigma notation is a for-loop that iterates m from 1 to M, accumulating the sum of the subfunction, fm, results. In the machine learning world, we're given a set of data points rathe