Embed Size (px)
DESCRIPTION
Transfer for Supervised Learning Tasks. HAITHAM BOU AMMAR MAASTRICHT UNIVERSITY. Motivation. Model. Assumptions: Training and Test are from same distribution Training and Test are in same feature space. Not True in many real-world applications. Training Data. ?. Test Data. - PowerPoint PPT Presentation
Citation preview
HAITHAM BOU AMMAR
MAASTRICHT UNIVERSITY
Transfer for Supervised Learning Tasks
Motivation
Model
€
y
€
x
€
x€
y
Training Data
?
Test Data
Assumptions: 1.Training and Test are from same distribution 2.Training and Test are in same feature spaceNo
t Tru
e in
man
y re
al-
world
app
licat
ions
Examples: Web-document Classification
Model
?
Physics Machine Learning
Life Science
Examples: Web-document Classification
Model
?
Physics Machine Learning
Life Science
Content Change !
Assumption violated!
Learn a new model
Learn new Model :
1.Collect new Labeled Data2.Build new model
Reuse & Adapt already learned model !
Examples: Image Classification
Model OneFeatures Task One
Task One
Examples: Image Classification
Cars
Motorcycles
Task Two
Features Task One
Features Task Two
Reuse
Model Two
Traditional Machine Learning vs. Transfer
Source Task
Knowledge
Target Task
Learning System
Different Tasks
Learning System
Learning System
Learning System
Traditional Machine Learning
Transfer Learning
Questions ?
Transfer Learning Definition
Notation: Domain :
Feature Space: Marginal Probability Distribution:
with
Given a domain then a task is :
Labe
l
Space P(Y
|X)
Transfer Learning Definition
Given a source domain and source learning task, a target domain and a target learning task, transfer learning aims to help improve the learning of the target predictive function using the source knowledge, where
or
Transfer Definition
Therefore, if either : Domain Differences
Task Differences
Examples: Cancer Data
Age Smoking
Age Height Smoking
Examples: Cancer Data
Task S
ourc
e: Clas
sify
into c
ance
r or n
o
canc
erTa
sk T
arge
t: Clas
sify
into c
ance
r lev
el on
e,
canc
er le
vel t
wo,
canc
er le
vel t
hree
Quiz
When doesn’t transfer help ? (Hint: On what should you
condition?)
Questions ?
Questions to answer when transferringW
hat t
o Tra
nsfe
r ?
How to
Trans
fer ?
Whe
n to
Tra
nsfe
r ?
Instan
ces
?
Model
?
Features
?
Map
M
odel
?
Unify
Feat
ures
?
Weig
ht
Insta
nces
?
In w
hich
Si
tuat
ions
Algorithms: TrAdaBoost
Assumptions: Source and Target task have same feature space:
Marginal distributions are different:
Not all source data might be helpful !
Algorithms: TrAdaBoost (Quiz)
Wha
t to
Trans
fer ?
How to
Tra
nsfer
?
Instan
ces
Weig
ht In
stanc
es
Algorithm: TrAdaBoost
Idea: Iteratively reweight source samples such that:
reduce effect of “bad” source instances encourage effect of “good” source instances
Requires: Source task labeled data set Very small Target task labeled data set Unlabeled Target data set Base Learner
Algorithm: TrAdaBoost
Weights Initialization
Hypothesis Learning and error calculation
Weights Update
Questions ?
Algorithms: Self-Taught Learning
Problem Targeted is :
1.Little labeled data
2.A lot of unlabeled data
Build a model on the labeled data
Natural scenes
Car
Motorcycl
e
Algorithms: Self-Taught Learning
Assumptions: Source and Target task have different feature space:
Marginal distributions are different:
Label Space is different:
Algorithms: Self-Taught Learning (Quiz)
Wha
t to
Trans
fer ?
How to
Tra
nsfer
?
1. Disc
over
Featu
res
2. Unif
y Fea
ture
s
Features
Algorithms: Self-Taught Learning
Framework: Source Unlabeled data set:
Target Labeled data set: Natural
scenes
Build classifier for cars and Motorbikes
Algorithms: Self-Taught Learning
Step One: Discover high level features from Source data by
Regularization Term Re-construction Error
Constraint on the Bases
Algorithm: Self-Taught Learning
Unlabeled Data Set
High
Leve
l Fea
ture
s
Algorithm: Self-Taught Learning
Step Two: Project target data onto the attained features by
Informally, find the activations in the attained bases such that: 1.Re-construction is minimized 2.Attained vector is sparse
Algorithms: Self-Taught Learning
High Level F
eatures
Algorithms: Self-Taught Learning
Step Three: Learn a Classifier with the new features
Target TaskSource Task
Learn new features (Step 1)
Project target data (Step 2)Learn Model (Step 3)
Conclusions
Transfer learning is to re-use source knowledge to help a target learner
Transfer learning is not generalization
TrAdaBoot transfers instances
Self-Taught learning transfer unlabeled features
hmm
