Submodular Bandits and their Application to Diversified Retrieval Yisong Yue CMU amp Carlos Guestrin CMU Optimizing Recommender Systems Every day users come to news portal For each user ID: 183052
Download Presentation The PPT/PDF document "Linear" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Linear
Submodular
Bandits
and their Application to Diversified Retrieval
Yisong Yue (CMU) & Carlos
Guestrin (CMU)
Optimizing Recommender Systems
Every day, users come to news portal
For each user,
News portal recommends L articles to cover the user’s interestsUsers provide feedback (clicks, ratings, “likes”).System integrates feedback for future use.
Challenge 1: Making Diversified Recommendations
Should recommend optimally diversified sets of articles.
“Israel implements unilateral Gaza cease-fire :: WRAL.com”“Israel unilaterally halts fire, rockets persist”“Gaza truce, Israeli pullout begin | Latest News”“Hamas announces ceasefire after Israel declares truce - …”“Hamas fighters seek to restore order in Gaza Strip - World - Wire …”
“Israel implements unilateral Gaza cease-fire :: WRAL.com”“Obama vows to fight for middle class”“Citigroup plans to cut 4500 jobs”“Google Android market tops 10 billion downloads”“UC astronomers discover two largest black holes ever found”
Challenge 2: Personalization
Modeling Diversity via
Submodular
Utility Functions
We assume a set of D concepts or topicsUsers are modeled by how interested they are in each topicLet Fi(A) denote the how well set of articles A covers topic i. (“topic coverage function”)We model user utility as F(A|w) = wT[F1(A), …, FD(A)]
Linear
Submodular Bandits Problem
At each iteration t:A set of available articles, AtEach article represented using D submodular basis functionsAlgorithm selects a set of L articles AtAlgorithm recommends At to user, receives feedbackAssumptions:Pr(like | a,A) = wTΔ(a|A) (conditional submodular independence)Regret: (1-1/e)OPT – sum of rewards
Goal: recommend a set of articles that optimally covers topics that interest the user.
Each topic coverage function Fi(A) is monotone submodular! A function F is submodular if i.e., the benefit of recommending a second (redundant) article is smaller than adding the first.
Properties of
Submodular
Functions
Sums of submodular functions are submodularSo F(A|w) is submodularExact inference is NP-hard!Greedy algorithm yields (1-1/e) approximation boundIncremental gains are locally linear!Both properties will be exploited by our online learning algorithm
We address two challenges:
Diversified recommendations
Exploration for personalization
Example: Probabilistic Coverage
LSBGreedy
News Recommender User Study
OR
?
Different users have different interests
Can only learn interests by recommending and receiving feedback
Exploration versus exploitation dilemma
We model this as a bandit problem!
Mean Estimate by Topic
Uncertainty of Estimate
10 days, 10 articles per day
Compared against
Multi.
Weighting
(no exploration
) [El-Arini et al, ‘09]
Ranked Bandits +
LinUCB
(reduction approach, does not directly model diversity) [
Radlinski
et al, ’08; Li et al., ‘10]
Comparing learned weights for two sessions (
LSBGreedy
vs
MW)
1
st
session, MW
overfits
to “world “ topic
2
nd
session, user liked few articles, and MW did not learn anything
Maintain mean and confidence interval of user’s interests
Greedily recommend articles with highest upper confidence utility
In example below, chooses article about economy
Theorem: with probability 1- δ average regret shrinks as
Comparison
Win / Tie / LossGain per Day % LikesLSBGreedy vs Static 24 / 0 / 0 1.07 67%LSBGreedy vs MW 24 / 1 / 1 0.54 63%LSBGreedy vs RankLinUCB 21 / 2 / 4 0.58 61%
+
Each article a has probability P(
i|a
) of covering topic I
Define topic coverage function for set A as
Straightforward to show that F is monotone
submodular
[El-
Arini
et al., ‘09]