Download presentation
Presentation is loading. Please wait.
Published byCaleb Keeble Modified over 10 years ago
1
Graphical Models for the Internet Amr Ahmed and Alexander Smola Yahoo Research, Santa Clara, CA
2
Thus far... Motivation Basic tools – Clustering – Topic Models Distributed batch inference – Local and global states – Star synchronization
3
Up next Inference – Online Distributed Sampling – Single machine multi-threaded inference – Online EM and Submodular Selection Applications – User tracking for behavioral Targeting – Content understanding – User modeling for content recommendation
4
4. Online Model
5
Scenarios Batch Large-Scale – Covered in part 1 Mini-batches – We already have a model – Data arrives in batches – We would like to keep model up-to-data Time-sensitive – Data arrives one item at a time – Model should be up-to-data Time
6
4.1 Dynamic Clustering
7
The Chinese Restaurant Process Allows the number of mixtures to grow with the data They are called non-parametric models – Means the number of effective parameters grow with data – Still have hyper-parameters that control the rate of growth : how fast a new cluster/mixture is born? G 0 : Prior over mixture component parameters
8
The Chinese Restaurant Process -For data point x i - Choose table j m j and Sample x i ~ f( j ) - Choose a new table K+1 - Sample K+1 ~ G 0 and Sample x i ~ f( K+1 ) Generative Process 2 1 3 The rich gets richer effect CANNOT handle sequential data
9
Recurrent CRP (RCRP) [Ahmed and Xing 2008] Adapts the number of mixture components over time – Mixture components can die out – New mixture components are born at any time – Retained mixture components parameters evolve according to a Markovian dynamics
10
The Recurrent Chinese Restaurant Process 2,1 1,1 3,1 T=1 Dish eaten at table 3 at time epoch 1 OR the parameters of cluster 3 at time epoch 1 -Customers at time T=1 are seated as before: - Choose table j m j,1 and Sample x i ~ f( j,1 ) - Choose a new table K+1 - Sample K+1,1 ~ G 0 and Sample x i ~ f( K+1,1 ) Generative Process
11
The Recurrent Chinese Restaurant Process 2,1 1,1 3,1 T=1 T=2 2,1 1,1 3,1 m' 1,1 =2 m' 2,1 =3m' 3,1 =1
12
2,1 1,1 3,1 T=1 T=2 2,1 1,1 3,1 m' 1,1 =2 m' 2,1 =3m' 3,1 =1
13
2,1 1,1 3,1 T=1 T=2 2,1 1,1 3,1 m' 1,1 =2 m' 2,1 =3m' 3,1 =1
14
2,1 1,1 3,1 T=1 T=2 2,1 1,1 3,1 m' 1,1 =2 m' 2,1 =3m' 3,1 =1
15
2,1 1,1 3,1 T=1 T=2 2,1 1,2 3,1 Sample 1,2 ~ P(.| 1,1 ) m' 1,1 =2 m' 2,1 =3m' 3,1 =1
16
2,1 1,1 3,1 T=1 T=2 2,1 1,2 3,1 And so on …… m' 1,1 =2 m' 2,1 =3m' 3,1 =1
17
2,1 1,1 3,1 T=1 T=2 2,2 1,2 3,1 At the end of epoch 2 4,2 Newly born cluster Died out cluster m' 1,1 =2 m' 2,1 =3m' 3,1 =1
18
2,1 1,1 3,1 T=1 T=2 2,2 1,2 3,1 N 1,1 =2 N 2,1 =3m' 3,1 =1 4,2 T=3 2,2 1,2 4,2 m' 4,2 =1m' 2,2 =2m' 1,2 =2
19
Recurrent Chinese Restaurant Process Can be extended to model higher-order dependencies Can decay dependencies over time – Pseudo-counts for table k at time t is History size Decay factory Number of customers sitting at table K at time epoch t-h H h htk h me 1,
20
2,1 1,1 3,1 T=1 T=2 2,2 1,2 3,1 m' 1,1 =2 m' 2,1 =3m' 3,1 =1 4,2 T=3 2,2 1,2 4,2 H h htk h me 1, m' 2,3 m' 2,3 =
21
4.2 Online Distributed Inference Tracking Users Interest
22
Characterizing User Interests Short term vs long-term Jan April July Oct Music Housing Furniture Buying a car Travel plans
23
Characterizing User Interests Short term vs long-term Latent Jan April July Oct millage fast mortgage Gaga seafood Barcelona used
24
Problem formulation Input Queries issued by the user or tags of watched content Snippet of page examined by user Time stamp of each action (day resolution) Output Users daily distribution over interests Dynamic interest representation Online and scalable inference Language independent Flight London Hotel weather School Supplies Loan semester classes registration housing rent
25
Problem formulation Flight London Hotel weather Travel Back To school finance School Supplies Loan semester classes registration housing rent Input Queries issued by the user or tags of watched content Snippet of page examined by user Time stamp of each action (day resolution) Output Users daily distribution over interests Dynamic interest representation Online and scalable inference Language independent
26
Problem formulation Flight London Hotel weather Travel Back To school finance School Supplies Loan semester classes registration housing rent When to show a financing ad?
27
Problem formulation Flight London Hotel weather Travel Back To school finance School Supplies Loan semester classes registration housing rent When to show a financing ad?
28
Problem formulation Flight London Hotel weather Travel Back To school finance School Supplies Loan semester classes registration housing rent When to show a financing ad?
29
Problem formulation Flight London Hotel weather Travel Back To school finance School Supplies Loan semester classes registration housing rent When to show a hotel ad?
30
Problem formulation Flight London Hotel weather Travel Back To school finance School Supplies Loan semester classes registration housing rent When to show a hotel ad?
31
Problem formulation Flight London Hotel weather Travel Back To school finance School Supplies Loan semester classes registration housing rent Input Queries issued by the user or tags of watched content Snippet of page examined by user Time stamp of each action (day resolution) Output Users daily distribution over interests Dynamic interest representation Online and scalable inference Language independent
32
Mixed-Membership Formulation job Career Business Assistant Hiring Part-time Receptionist job Career Business Assistant Hiring Part-time Receptionist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder Diet Cars Job Finance Objects Mixtures Degree of membership Job Hiring speed price part-time Camry Career opening bonus package Job Hiring speed price part-time Camry Career opening bonus package card diet calories loan recipe milk Weight lb kg card diet calories loan recipe milk Weight lb kg
33
In Graphical Notation
34
In Polya-Urn Representation Collapse multinomial variables: Fixed-dimensional Hierarchal Polya-Urn representation – Chinese restaurant franchise x x
35
Food Chicken job Career Business Assistant Hiring Part- time Receptio nist job Career Business Assistant Hiring Part- time Receptio nist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder Car speed offer camry accord career Global topics trends Topic word-distributions Topic word-distributions User-specific topics trends (mixing-vector) User-specific topics trends (mixing-vector) User interactions: queries, keyword from pages viewed
36
Food Chicken Car speed offer camry accord career For each user interaction Choose an intent from local distribution Sample word from the topics word-distribution Choose a new intent Sample a new intent from the global distribution Sample word from the new topic word- distribution Generative Process ……… job Career Business Assistant Hiring Part- time Receptio nist job Career Business Assistant Hiring Part- time Receptio nist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder
37
Food Chicken pizza Car speed offer camry accord career For each user interaction Choose an intent from local distribution Sample word from the topics word-distribution Choose a new intent Sample a new intent from the global distribution Sample word from the new topic word- distribution Generative Process ……… job Career Business Assistant Hiring Part- time Receptio nist job Career Business Assistant Hiring Part- time Receptio nist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder
38
Food Chicken pizza Car speed offer camry accord career For each user interaction Choose an intent from local distribution Sample word from the topics word-distribution Choose a new intent Sample a new intent from the global distribution Sample word from the new topic word- distribution Generative Process ……… job Career Business Assistant Hiring Part- time Receptio nist job Career Business Assistant Hiring Part- time Receptio nist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder
39
Food Chicken pizza hiring Car speed offer camry accord career For each user interaction Choose an intent from local distribution Sample word from the topics word-distribution Choose a new intent Sample a new intent from the global distribution Sample word from the new topic word- distribution Generative Process ……… job Career Business Assistant Hiring Part- time Receptio nist job Career Business Assistant Hiring Part- time Receptio nist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder
40
Food Chicken pizza millage Car speed offer camry accord career For each user interaction Choose an intent from local distribution Sample word from topics word-distribution Choose a new intent Sample a new intent from the global distribution Sample from word the new topic word- distribution Generative Process job Career Business Assistant Hiring Part- time Receptio nist job Career Business Assistant Hiring Part- time Receptio nist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder
41
Food Chicken pizza millage Car speed offer camry accord career job Career Business Assistant Hiring Part- time Receptio nist job Career Business Assistant Hiring Part- time Receptio nist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder Problems - Static Model - Does not evolve users interests - Does not evolve the global trend of interests - Does not evolve interests distribution over terms
42
Food Chicken pizza millage Car speed offer camry accord career job Career Business Assistant Hiring Part- time Receptio nist job Career Business Assistant Hiring Part- time Receptio nist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder At time tAt time t+1 Build a dynamic model Connect each level using a RCRP Connect each level using a RCRP
43
Which time kernel to use at each level? Which time kernel to use at each level?
44
Pseudo counts = * Decay factor Food Chicken pizza millage Car speed offer camry accord career At time t Observation 1 -Popular topics at time t are likely to be popular at time t+1 k,t+1 is likely to smoothly evolve from k,t At time t+1 job Career Business Assistant Hiring Part- time Receptio nist job Career Business Assistant Hiring Part- time Receptio nist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder
45
Food Chicken pizza millage Car speed offer camry accord career At time tAt time t+1 job Career Business Assistant Hiring Part- time Receptio nist job Career Business Assistant Hiring Part- time Receptio nist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder Observation 1 -Popular topics at time t are likely to be popular at time t+1 k,t+1 is likely to smoothly evolve from k,t Car Altima Accord Book Kelley Prices Small Speed Car Altima Accord Book Kelley Prices Small Speed k,t+1 Dir k,t+1 k,t Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Intuition Captures current trend of the car industry (new release for e.g.) ~
46
Food Chicken pizza millage Car speed offer camry accord career At time tAt time t+1 Observation 2 - User prior at time t+1 is a mixture of the user short and long term interest How do we get a prior that captures both long and short term interest? job Career Business Assistant Hiring Part- time Receptio nist job Career Business Assistant Hiring Part- time Receptio nist Car Altima Accord Blue Book Kelley Prices Small Speed Car Altima Accord Blue Book Kelley Prices Small Speed Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder
47
All week job Career Business Assistant Hiring Part-time Receptionist job Career Business Assistant Hiring Part-time Receptionist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder month Time t t+1 Food Chicken pizza recipe job hiring Part-time Opening salary food chicken Pizza millage Kelly recipe cuisine Diet Cars Job Finance Prior for user actions at time t μ μ2μ2 μ3μ3 Long-term short-term
48
Food Chicken Pizza millage Car speed offer camry accord career At time tAt time t+1 For each user interaction Choose an intent from local distribution Sample word from the topics word-distribution Choose a new intent Sample a new intent from the global distribution Sample word from the new topic word- distribution Generative Process short-term priors job Career Business Assistant Hiring Part- time Receptio nist job Career Business Assistant Hiring Part- time Receptio nist Car Altima Accord Blue Book Kelley Prices Small Speed Car Altima Accord Blue Book Kelley Prices Small Speed Bank Online Credit Card debt portfolio Finance Chase Bank Online Credit Card debt portfolio Finance Chase Recipe Chocolate Pizza Food Chicken Milk Butter Powder Recipe Chocolate Pizza Food Chicken Milk Butter Powder
49
Polya-Urn RCRF Process Polya-Urn RCRF Process ? ?
50
Simplified Graphical Model ~ ~ ~ ~ ~ ~ At time tAt time t+1
51
Simplified Graphical Model ~ ~ ~ ~ ~ ~ Car Altima Accord Book Kelley Prices Small Speed Car Altima Accord Book Kelley Prices Small Speed Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large
52
Simplified Graphical Model ~ ~ ~ ~ ~ ~ Food Chicken Pizza millage
53
Simplified Graphical Model ~ ~ ~ ~ ~ ~
54
~ ~ ~ ~ ~ ~ Topics evolve over time? Users intent evolve over time? Capture long and term interests of users?
55
User 1 process User 2 process User 3 process Global process
56
4.2 Online Distributed Inference Work Flow
57
today Current Users models Current Users models System state System state new Users models new Users models User interactions Daily Update (inference) Daily Update (inference) Hundred of millions
58
~ ~ ~ ~ ~ ~ Online Scalable Inference Online algorithm – Greedy 1-particle filtering algorithm – Works well in practice – Collapse all multinomials except Ω t This makes distributed inference easier – At each time t: Distributed scalable implementation – Used first part architecture as a subroutine – Added synchronous sampling capabilities
59
Distributed Inference (at time t) z z w w
60
client Distributed Inference (at time t) z z w w z z w w Collapse all multinomial
61
After collapsing Food Chicken Pizza millage Food Chicken Pizza millage job Career Busines s Assista nt Hiring Part- time Recepti onist job Career Busines s Assista nt Hiring Part- time Recepti onist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfol io Financ e Chase Bank Online Credit Card debt portfol io Financ e Chase Recipe Chocol ate Pizza Food Chicke n Milk Butter Powde r Recipe Chocol ate Pizza Food Chicke n Milk Butter Powde r client Car speed offer camry accord career Car speed offer camry accord career Use Star-Synchronization
62
Fully Collapsed Food Chicken Pizza millage Food Chicken Pizza millage job Career Busines s Assista nt Hiring Part- time Recepti onist job Career Busines s Assista nt Hiring Part- time Recepti onist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfol io Financ e Chase Bank Online Credit Card debt portfol io Financ e Chase Recipe Chocol ate Pizza Food Chicke n Milk Butter Powde r Recipe Chocol ate Pizza Food Chicke n Milk Butter Powde r Shared memory client Car speed offer camry accord career Car speed offer camry accord career
63
Distributed Inference (at time t) Food Chicken Pizza millage Food Chicken Pizza millage client Car speed offer camry accord career Car speed offer camry accord career Local trend Global trend Topic factor
64
Semi-Collapsed Food Chicken Pizza millage Food Chicken Pizza millage job Career Busines s Assista nt Hiring Part- time Recepti onist job Career Busines s Assista nt Hiring Part- time Recepti onist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfol io Financ e Chase Bank Online Credit Card debt portfol io Financ e Chase Recipe Chocol ate Pizza Food Chicke n Milk Butter Powde r Recipe Chocol ate Pizza Food Chicke n Milk Butter Powde r Shared memory client Car speed offer camry accord career Car speed offer camry accord career
65
Semi-Collapsed Food Chicken Pizza millage Food Chicken Pizza millage job Career Busines s Assista nt Hiring Part- time Recepti onist job Career Busines s Assista nt Hiring Part- time Recepti onist Car Blue Book Kelley Prices Small Speed large Car Blue Book Kelley Prices Small Speed large Bank Online Credit Card debt portfol io Financ e Chase Bank Online Credit Card debt portfol io Financ e Chase Recipe Chocol ate Pizza Food Chicke n Milk Butter Powde r Recipe Chocol ate Pizza Food Chicke n Milk Butter Powde r Shared memory client ΩtΩt ΩtΩt ΩtΩt Car speed offer camry accord career Car speed offer camry accord career
66
Semi-Collapsed Food Chicken Pizza millage Food Chicken Pizza millage client ΩtΩt ΩtΩt Car speed offer camry accord career Car speed offer camry accord career
67
Distributed Sampling Cycle Sample Z For users Sample Z For users Sample Z For users Sample Z For users Sample Z For users Sample Z For users Sample Z For users Sample Z For users Barrier Write counts to memcached Write counts to memcached Write counts to memcached Write counts to memcached Write counts to memcached Write counts to memcached Write counts to memcached Write counts to memcached Collect counts and sample Do nothing Barrier Read from memcached Sample Ω t Requires a reduction step
68
Distributed Sampling Cycle Sample Z For users Sample Z For users Sample Z For users Sample Z For users Sample Z For users Sample Z For users Sample Z For users Sample Z For users Barrier Write counts Collect counts and sample Do nothing Barrier Read from
69
Sampling Introduce auxiliary variable m kt – How many times the global distribution was visited – ~ AnotniaK m m
70
Distributed Sampling Cycle Sample Z For users Sample Z For users Sample Z For users Sample Z For users Sample Z For users Sample Z For users Sample Z For users Sample Z For users Barrier Write counts Collect counts and sample Do nothing Barrier Read from
71
4.2 Online Distributed Inference Behavioral Targeting
72
Tasks is predicting convergence in display advertising Use two datasets – 6 weeks of user history – Last week responses to Ads are used for testing Baseline: – User raw data as features – Static topic model Experimental Results
73
Interpretability User-1User-2
74
Performance in Display Advertising ROC Number of conversions
75
Performance in Display Advertising Weighted ROC measure Static Batch models Static Batch models Effect of number of topics
76
How Does It Scale? 2 Billion instances with 5M vocabulary using 1000 machines one iteration took ~ 3.8 minutes 2 Billion instances with 5M vocabulary using 1000 machines one iteration took ~ 3.8 minutes
77
Distributed Inference Revisited
78
To collapse or not to collapse? Not collapsing – Keeps conditional independence Good for parallelization Requires synchronous sampling – Might mix slowly Collapsing – Mixes faster – Hinder parallelism – Use star-synchronization Works well if sibling depends on each others via aggregates Requires asynchronous communication
79
Inference Primitive Collapse a variable – Star synchronization for the sufficient statistics Sampling a variable – Local Sample it locally (possibly using the synchronized statistics) – Shared Synchronous sampling using a barrier Optimizing a variable – Same as in the shared variable case – Ex. Conditional topic models
80
Online Models Batch Large-Scale – Covered in part 1 Mini-batches – We already have a model – Data arrives in batches – We would like to keep model up-to-data Time-sensitive – Data arrives one item at a time – Model should be up-to-data Time
81
What Is Coming? Inference – Online Distributed Sampling – Single machine multi-threaded inference – Online EM and Submodular Selection Applications – User tracking for behavioral Targeting – Content understanding – User modeling for content recommendation
82
4.2 Scalable SMC Inference Storylines
83
News Stream
84
Realtime news stream – Multiple sources (Reuters, AP, CNN,...) – Same story from multiple sources – Stories are related Goals – Aggregate articles into a storyline – Analyze the storyline (topics, entities) How does the story develop over time? Who are the main entities?
85
A Unified Model Jointly solves the three main tasks – Clustering, – Classification – Analysis Building blocks – A Topic model High-level concepts (unsupervised classification) – Dynamic clustering (RCRP) Discover tightly-focused concepts – Named entities – Story developments
86
Infinite Dynamic Cluster-Topic Hybrid Sports games Won Team Final Season League held Politics Government Minister Authorities Opposition Officials Leaders group Unrest Police Attack run man group arrested move UEFA-soccer Champions Goal Coach Striker Midfield penalty Juventus AC Milan Lazio Ronaldo Lyon Tax-Bill Tax Billion Cut Plan Budget Economy Bush Senate Fleischer White House Republican
87
Infinite Dynamic Cluster-Topic Hybrid Sports games Won Team Final Season League held Politics Government Minister Authorities Opposition Officials Leaders group Unrest Police Attack run man group arrested move UEFA-soccer Champions Goal Coach Striker Midfield penalty Juventus AC Milan Lazio Ronaldo Lyon Tax-Bill Tax Billion Cut Plan Budget Economy Bush Senate Fleischer White House Republican Border-Tension Nuclear Border Dialogue Diplomatic militant Insurgency missile Pakistan India Kashmir New Delhi Islamabad Musharraf Vajpayee
88
The Graphical Model Topic model Topics per cluster RCRP for cluster Hierarchical DP for article Separate model for named entities Story specific correction
89
4.2 Fast SMC Inference Inference via SMC
90
Online Inference Algorithm A Particle filtering algorithm Each particle maintains a hypothesis – What are the stories – Document-story associations – Topic-word distributions Collapsed sampling – Sample (z d,s d ) only for each document
91
Particle Filter Representation
92
Fold the document into the structure of each filter -s and z are tightly coupled -Alternatives -Sample s then sample z (high variance) wwwwwwentities Document td zzzzzz s
93
Fold the document into the structure of each filter -s and z are tightly coupled -Alternatives -Sample s then sample z (high variance) -Sample z then sample s (doesnt make sense) wwwwwwentities Document td zzzzzz s
94
Fold the document into the structure of each filter -s and z are tightly coupled -Alternatives -Sample s then sample z (high variance) -Sample z then sample s (doesnt make sense) -Idea -Run a few iterations of MCMC over s and z -Take last sample as the proposed value
95
How good each filter look now?
96
Get rid of bad filter Replicate good one
97
Get rid of bad filter Replicate good one
98
Efficient Computation and Storage Particles get replicated 1 State P1 12 State P2 State P1 12 3 State P2 State P3 State P1
99
Efficient Computation and Storage Particles get replicated – Use thread-safe Inheritance tree 1 State P1 12 State P2 State P1 12 3 State P2 State P3 State P1
100
Efficient Computation and Storage Particles get replicated – Use thread-safe Inheritance tree [extends Canini et. Al 2009] 1 state 1
101
Efficient Computation and Storage Particles get replicated – Use thread-safe Inheritance tree [extends Canini et. Al 2009] Root state 2 1 State 1 12
102
Efficient Computation and Storage Particles get replicated – Use thread-safe Inheritance tree [extends Canini et. Al 2009] Root 3 state 1 2 State state State 1 12 12 3
103
Efficient Computation and Storage Particles get replicated – Use thread-safe Inheritance tree [extends Canini et. Al 2009] – Inverted representation for fast lookup Root 3 India: story 5 Pakistan: story 1 1 2 (empty)Congress: story 2 Bush: story 2 India: story 3 India: story 1, US: story 2, story 3 1 12 12 3
104
Efficient Computation and Storage Why this is useful? Only focus on stories that mention at least one entity – Otherwise pre-compute and reuse We can use fast samplers for z as well [Yao et. Al. KDD09]
105
Experiments Yahoo! News datasets over two months – Three sub-sampled sets with different characteristics Editorially-labeled documents – Cannot-like and must-link pairs Performance measures using clustering accuracy Baseline – A strong offline Correlation clustering algorithm [WSDM 11] Scaled with LSH to compute neighborhood graph (similar to Petrovic 2010)
106
Structured Browsing Sports games Won Team Final Season League held Politics Government Minister Authorities Opposition Officials Leaders group Unrest Police Attach run man group arrested move Border-Tension Nuclear Border Dialogue Diplomatic militant Insurgency missile Pakistan India Kashmir New Delhi Islamabad Musharraf Vajpayee UEFA-soccer Champions Goal Leg Coach Striker Midfield penalty Juventus AC Milan Real Madrid Milan Lazio Ronaldo Lyon Tax-bills Tax Billion Cut Plan Budget Economy lawmakers Bush Senate US Congress Fleischer White House Republican
107
Structured Browsing More Like India-Pakistan story Middle-east-conflict Peace Roadmap Suicide Violence Settlements bombing Israel Palestinian West bank Sharon Hamas Arafat Based on topics Nuclear programs Nuclear summit warning policy missile program North Korea South Korea U.S Bush Pyongyang Nuclear+ topics [politics] Border-Tension Nuclear Border Dialogue Diplomatic militant Insurgency missile Pakistan India Kashmir New Delhi Islamabad Musharraf Vajpayee
108
Structured Browsing Sports games Won Team Final Season League held Politics Government Minister Authorities Opposition Officials Leaders group Unrest Police Attach run man group arrested move Border-Tension Nuclear Border Dialogue Diplomatic militant Insurgency missile Pakistan India Kashmir New Delhi Islamabad Musharraf Vajpayee UEFA-soccer Champions Goal Leg Coach Striker Midfield penalty Juventus AC Milan Real Madrid Milan Lazio Ronaldo Lyon Tax-bills Tax Billion Cut Plan Budget Economy lawmakers Bush Senate US Congress Fleischer White House Republican More on Personalization later on the talk More on Personalization later on the talk
109
Quantitative Evaluation Number of topics = 100 Effect of number of topics
110
Scalability
111
Model Contribution Named entities are very important Removing time increase processing up to 2 seconds per document
112
Putting Things Together
113
Time vs. Machines Data arrives dynamically How to keep models up to date? BatchMini-batchesTruly online Single-Machine Gibbs Variational Online-LDASMC Multi-MachineStar-Synch.Star-Synch + Synchronous step ?
114
4.3 User Preference Online EM and Submodularity
115
Storyline Summarization How to summarize a storyline with few articles? Earthquake & Tsunami Nuclear Power Rescue & ReliefEconomy Time
116
Storyline Summarization Earthquake & Tsunami Nuclear Power Rescue & ReliefEconomy Time How to summarize a storyline with few articles? How to personalize the summary?
117
Storyline Summarization Earthquake & Tsunami Nuclear Power Rescue & ReliefEconomy Time How to summarize a storyline with few articles? How to personalize the summary?
118
User Interaction Passive – We observe the user generated contents – Model user based on those content using unsupervised techniques Explicit – We present users with content – User give explicit feedback Like/dislike – Learn user preference using supervised techniques Implicit – Mixture between the two – Present the user with items – Observe which items the user interact with – Learning user preference using semi-supervised models
119
User Satisfaction Modular – Present users with items she prefers Regardless of the context – Targets relevance – Ex: vector space models Submodular – More of the same thing is not always better Dimensioning return – Targets diversity – Ex: TDN [ElArini et. Al. KDD 09]
120
Sequential Click-View Model Modeling Views based on position
121
Sequential Click-View Model Threshold Information gain #clicks Modeling clicks using position and information gain Coverage functions weights are learnt
122
Sequential Click-View Model Selected Summary Story Features Modular Submodula r
123
Online Inference Treat missing views as hidden variables – Realistic interaction model Use the online EM algorithm – Infer the value of hidden variables Optimize parameters using SGD – Use additive weights Background + story + category + user
124
Online Inference
125
How Does it Work?
126
How Does It Work?
127
5. Summary Future Directions
128
Summary Tools – Load distribution, balancing and synchronization – Clustering, Topic Models Models – Dynamic non-parametric models – Sequential latent variable models Inference Algorithms – Distributed batch – Sequential Monte Carlo Applications – User profiling – News content analysis & recommendation
129
Future Directions Theoretical bounds and guarantees Network data – Graph partitioning Non-parametric models – Learning structure from data Working under communication constraints Data distribution for particle filters
Similar presentations
© 2024 SlidePlayer.com Inc.
All rights reserved.