Professional Documents
Culture Documents
(_hx“)
Lecture 2: The Learning Problems
Hsuan-Tien Lin (ó“0)
htlin@csie.ntu.edu.tw
探討 不同 的 ML 問題 Roadmap
1 When Can Machines Learn?
hypothesis set
H
• credit approve/disapprove
• email spam/non-spam
• patient sick/not sick
• ad profitable/not profitable
• answer correct/incorrect (KDDCup 2010)
5
• Y = {1c, 5c, 10c, 25c}, or
1 Y = {1, 2, · · · , K } (abstractly) 多 選題
• binary classification: special case
10
with K = 2
Size
?
(image by Robert-Owen-Wahl from Pixabay)
Which Fruits?
multilabel classification: 複選 題
classify input to multiple (or no) categories
Y = 2{apple,orange,strawberry,kiwi}
powerset ,
What Tags?
+ )
(Leonardo da Vinci, (Van Gogh, (Pjfinlay,
in Public Domain) in Public Domain) with CC0)
all images are downloaded from Wikipedia
Mini Summary
Learning with Different Output Space Y
是非 題 • binary classification: Y = { 1, +1}
選擇題 • multiclass classification: Y = {1, 2, · · · , K }
4逾
4選 … ⼼ 多 選題 • multilabel classification: Y = 2{1,2,··· ,K }
• regression: Y = R power 包含 所有 可能 的 了
比
hypothesis set
H
Questions?
Mass
5
unknown target function 1
f: X !Y
10
Size
hypothesis set
H
supervised learning:
every xn comes with corresponding yn
25
Mass
Mass
5
1
10
Size Size
25
Mass
Mass
5
1
10
Size Size
25 25
Mass
Mass
Mass
5 5
1 1
10 10
Reinforcement Learning
a ‘very different’ but natural way of learning
Reinforcement Learning
a ‘very different’ but natural way of learning
(Public Domain, from Wikipedia; used here for education purpose; all other rights still belong to Google DeepMind)
(CC-BY-SA 3.0 by Stannered on (CC-BY-SA 2.0 by Frej Bjon on (Public Domain, from Wikipedia)
Wikipedia) Wikipedia)
• weakly-supervised: no true yn
unknown target function
f: X !Y
• reinforcement: implicit yn by goodness(ỹn )
•
• . . . and more!!
hypothesis set
H
Questions?
Mass
5
unknown target function 1
f: X !Y
10
Size
hypothesis set
H
Mass
Mass
5
1
10
Size Size
real-world ML system
different from textbook settings
Hsuan-Tien Lin (NTU CSIE) Machine Learning 26/39
The Learning Problems Learning with Different Protocol f ) (xn , yn )
hypothesis set
H
Mini Summary
Learning with Different Protocol f ) (xn , yn )
• batch: all known data Offline =
hypothesis set
H
Questions?
hypothesis set
H
Mass
• patient info for cancer diagnosis 5
1
• often including ‘human intelligence’
on the learning task 10
Size
wronglabeimy.e
• 16 by 16 gray image x ⌘
(0, 0, 0.9, 0.6, · · · ) 2 R256
• ‘simple physical meaning’;
thus more difficult for ML
than concrete features
x =(symmetry, density)
1 較好 觕 5 較 dense 、
concrete ⼀些
变 成 有意✗ data is it a ‘1’ ? - z1 z5 is it a ‘5’ ?
patches
lmoreconcrete)
2 3 4 5
1 6
>
onekindoffeatnre
raw
, eytractfromimagei
• layered extraction: simple to complex features
• natural for difficult learning task with raw features, like vision
Mini Summary
Learning with Different Input Space X
• concrete: sophisticated (and related)
physical meaning
• raw: simple physical meaning
unknown target function
f: X !Y • abstract: no (or little) physical meaning
• . . . and more!!
hypothesis set
H
Questions?
Summary
1 When Can Machines Learn?