Escolar Documentos
Profissional Documentos
Cultura Documentos
Ralf Herbrich
Amazon
Overview
Theory
Inference in Factor Graphs
Approximate Message Passing
Applications @ Microsoft
TrueSkill: Gamer Rating and Matchmaking
TrueSkill Through Time: History of Chess
Click-Through Rate Prediction in Online Advertising
Matchbox: Recommendation Systems
Applications @ Amazon
Background Material
http://www.coursera.org http://www.cs.ubc.ca/~murphyk/MLbook/index.html
http://www.cs.ucl.ac.uk/staff/d.barber/brml/ http://research.microsoft.com/en-us/um/people/cmbishop/PRML/index.htm
Overview
Theory
Inference in Factor Graphs
Approximate Message Passing
Applications
TrueSkill: Gamer Rating and Matchmaking
TrueSkill Through Time: History of Chess
Click-Through Rate Prediction in Online Advertising
Matchbox: Recommendation Systems
Future Applications
Graphical Models
Semantic: a b
c
Local variable dependency of factors
Factor Graphs and Bayes Law
Bayes law
s1 s s2
Factorising prior
t1 t2
Factorising likelihood
d
Inference: Sum out latent variables
y
Factor Trees: Separation
y
f3(x,y)
v w x
f1(v,w) f2(w,x)
z
f4(x,z)
f2(w,x)
z
f4(x,z)
f2(w,x)
z
f4(x,z)
Redundant computations:
t
* =
-5 0 5 -5 0 5 -5 0 5
* =
-5 0 5 -5 0 5 -5 0 5
Divergence Measures
Kullback-Leibler Divergence: Expected log-odd ratio
between two distributions:
Special Cases:
-Divergence in Pictures
When to use which -Divergence?
x y
w1 w2
Theory
Inference in Factor Graphs
Approximate Message Passing
Applications @ Microsoft
TrueSkill: Gamer Rating and Matchmaking
TrueSkill Through Time: History of Chess
Click-Through Rate Prediction in Online Advertising
Matchbox: Recommendation Systems
Applications @ Amazon
TrueSkill
Joint work with Thore Graepel, Tom Minka & Phillip Trelford
Motivation
Given:
Match outcomes: Orderings among k teams
consisting of n1, n2 , ..., nk players, respectively
Questions:
Skill si for each player such that
s1 s2
p1 p2
y12
Two Team Match Outcome Model
s1 s2 s3 s4
t1 t2
y12
Multiple Team Match Outcome Model
s1 s2 s3 s4
t1 t2 t3
y
Multiple Team Match Outcome Model
s1 s2 s3 s4
t1 t2 t3
y12 y23
Efficient Approximate Inference
s1 s2 s3 s4
Leaderboard
Global ranking of all players
Matchmaking
For gamers: Most uncertain outcome
For inference: Most informative
Both are equivalent!
Experimental Setup
40
35
30
25
Level
20
15
char (TrueSkill)
10
SQLWildman (TrueSkill)
5 char (Halo 2 rank)
SQLWildman (Halo 2 rank)
0
0 100 200 300 400
Number of Games
Convergence Speed (ctd.)
100%
char wins
SQLWildman wins
Winning probability
60%
40%
20%
5/8 games won by char
0%
0 100 200 300 400 500
Garry Kasparov
3000
Mikhail Botvinnik
Skill estimate
1800
Jose Raul Capablanca
1600
Adolf Anderssen
1400
1850 1858 1866 1875 1883 1891 1899 1907 1916 1924 1932 1940 1949 1957 1965 1973 1981 1990 1998 2006
Year
Online
Advertising
Joint work with Thore Graepel, Joaquin Quionero Candela, Onno Zoeter, Tom Borchert , Phillip Trelford
Why Predict Probability-of-Click?
Display (according to
expected revenue)
Charge (per click)
$1.00 * 10% =$0.10 $0.80
Advantages
$2.00 of improved
* 4% =$0.08 probability
$1.25 estimates:
Increase$0.10
user satisfaction
* 50% =$0.05 by better
$0.05targeting
15.70.165.9
Client IP
221.98.2.187
92.154.3.86
+ p(pClick)
Match Exact Match
Type Broad Match
ML-1
Position SB-1
SB-2
Training Algorithm in Action
w1 + w2
c
No Click
Prediction
Training/Update
Click
Inference: An Optimization View
Accuracy
MatchBox
Joint work with Thore Graepel, Joaquin Quionero Candela, David Stern, Ulrich Paquet
Crime Drama Action Comedy Action Action
1 2 3 4 5 6
Programmer
Age<30
A
ID=33451
Student
Age<30
ID=33431
B
Shopkeeper
Age>45
C
ID=4321
Student
Age<30 D
ID=5641
Matchbox With Metadata
User Metadata Item Metadata
ID=234 Male British Camera SLR
u01 u11 u21 User v11 v21
+ s2 User trait 2 t2 +
Rating potential ~
r
Recommender System: MatchBox
User
likes
dislikes Social Network
Movie
Movie
mark Heat
sheryl
R. Scott
Director
Gender
Male C. Eastwood
Female Q. Tarantino
R. Howard
Message Passing For Matchbox
+ s1 * t1 +
+ s2 * t2 +
r
1.5
User/Item Trait Space
24: Season 3 Adaptation
1
24: Season 2
0.5
A Knights Tale
-0.5
-1
Users
A Cinderella Story Movies
-1.5
Incremental Training with ADF
Items
1 2 3 4 5 6
B
Users
D
ADF: Message Passing Iteration 1
1.5
0.5
0
-1.5 -1 -0.5 0 0.5 1 1.5
-0.5
-1
-1.5
Message Passing Iteration 2
1.5
0.5
0
-1.5 -1 -0.5 0 0.5 1 1.5
-0.5
-1
-1.5
Message Passing Iteration 3
1.5
0.5
0
-1.5 -1 -0.5 0 0.5 1 1.5
-0.5
-1
-1.5
Message Passing Iteration 4
1.5
0.5
0
-1.5 -1 -0.5 0 0.5 1 1.5
-0.5
-1
-1.5
feedback models
Feedback Models
+ s1 t1 +
+ s2 t2 +
r
Feedback Models
+ s1 t1 +
+ s2 t2 +
r
Feedback Models
=3
Feedback Models
t0 t1 t2 t3
Feedback Models
>0
Message Passing: Compositionality
u11 u21 v11 v21
+ s1 t1 +
+ s2 t2 +
User Model Item Model
*
x1 x2 x3 x4 r
Context Model +
Theory
Inference in Factor Graphs
Approximate Message Passing
Applications @ Microsoft
TrueSkill: Gamer Rating and Matchmaking
TrueSkill Through Time: History of Chess
Click-Through Rate Prediction in Online Advertising
Matchbox: Recommendation Systems
Applications @ Amazon
ML Opportunities @ Amazon
89
XRay
Machine Translation
Machine Translation: Deep Dive
Language Translation
Model Model