Filters








40 Hits in 5.9 sec

Personalized click shaping through lagrangian duality for online recommendation

Deepak Agarwal, Bee-Chung Chen, Pradheep Elango, Xuanhui Wang
2012 Proceedings of the 35th international ACM SIGIR conference on Research and development in information retrieval - SIGIR '12  
Most of existing works rely on online user feedback, notably clicks, as the objective and maximize it by showing articles with highest click-through rates.  ...  In this paper, we tackle the challenge by proposing personalized click shaping.  ...  In this paper, we propose personalized click shaping which combines deep personalization with click shaping for multiobjective online content recommendation.  ... 
doi:10.1145/2348283.2348350 dblp:conf/sigir/AgarwalCEW12 fatcat:bpz6eazus5esbjeeqpdkckteda

Constrained Optimization for Homepage Relevance

Deepak Agarwal, Shaunak Chatterjee, Yang Yang, Liang Zhang
2015 Proceedings of the 24th International Conference on World Wide Web - WWW '15 Companion  
Since there are multiple widgets to rank when a user visits, launching a personalized model to simply optimize user engagement such as clicks is often inappropriate.  ...  We show through both offline experiments and online A/B tests that our optimization framework can lead to significant improvement in user engagement while satisfying the desired set of business objectives  ...  Recommending the best links personalized for each user is an important problem.  ... 
doi:10.1145/2740908.2745398 dblp:conf/www/AgarwalCYZ15 fatcat:dsdk727lh5b7bp4tpzjoggn6nm

A Real-Time Whole Page Personalization Framework for E-Commerce [article]

Aditya Mantha, Anirudha Sundaresan, Shashank Kedia, Yokila Arora, Shubham Gupta, Gaoyang Wang, Praveenkumar Kanumala, Stephen Guo, Kannan Achan
2020 arXiv   pre-print
E-commerce platforms consistently aim to provide personalized recommendations to drive user engagement, enhance overall user experience, and improve business metrics.  ...  Given varied user preferences, optimizing the placement of these carousels is critical for improved user satisfaction.  ...  Multi-objective optimization problems in personalized click shaping has also been tackled utilizing a Lagrangian duality formulation [13] .  ... 
arXiv:2012.04681v1 fatcat:jvnpswxqdvaz7ei7en67ovrkdm

Online Evaluation for Information Retrieval

Katja Hofmann, Lihong Li, Filip Radlinski
2016 Foundations and Trends in Information Retrieval  
Personalized click shaping through Lagrangian duality for online recommendation.  ...  For example, early work on using historical information for estimating online performance focused on ad placement [Langford et al., 2008] and news recommendation .  ... 
doi:10.1561/1500000051 fatcat:gzskufswxjdpxe2ordm223vy7e

Provable Convex Co-clustering of Tensors [article]

Eric C. Chi and Brian R. Gaines and Will Wei Sun and Hua Zhou and Jian Yang
2020 arXiv   pre-print
Finally, we apply the CoCo estimator to the cluster analysis of advertisement click tensor data from a major online company.  ...  Cluster analysis is a fundamental tool for pattern discovery of complex heterogeneous data.  ...  Acknowledgments The authors thank Xu Han for his help with the simulation experiments during the revision of this work.  ... 
arXiv:1803.06518v2 fatcat:tobmphiumjdvbbp7jj37y6aixa

Provable Convex Co-clustering of Tensors

Eric C Chi, Brian R Gaines, Will Wei Sun, Hua Zhou, Jian Yang
2020 Journal of machine learning research  
Finally, we apply the CoCo estimator to the cluster analysis of advertisement click tensor data from a major online company.  ...  Cluster analysis is a fundamental tool for pattern discovery of complex heterogeneous data.  ...  Acknowledgments The authors thank Xu Han for his help with the simulation experiments during the revision of this work.  ... 
pmid:33312074 pmcid:PMC7731944 fatcat:bjrmicvonfcupkwzctaprj44tq

Email Volume Optimization at LinkedIn

Rupesh Gupta, Guanfeng Liang, Hsiao-Ping Tseng, Ravi Kiran Holur Vijay, Xiaoyu Chen, Romer Rosales
2016 Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '16  
system and experimental results from online A/B tests.  ...  Online social networking services distribute various types of messages to their members.  ...  Agarwal et al. have employed multi-objective optimization for personalized click shaping in online content recommendation in [8] and for displaying promotional widgets on web applications under business  ... 
doi:10.1145/2939672.2939692 dblp:conf/kdd/GuptaLTVCR16 fatcat:fo672we46jdevjxgxnl6dsewf4

Optimization with Non-Differentiable Constraints with Applications to Fairness, Recall, Churn, and Other Goals [article]

Andrew Cotter, Heinrich Jiang, Serena Wang, Taman Narayan, Maya Gupta, Seungil You, Karthik Sridharan
2018 arXiv   pre-print
Furthermore, if the constraints are non-differentiable, then one cannot optimize the Lagrangian with gradient-based methods. To solve these issues, we introduce the proxy-Lagrangian formulation.  ...  This new formulation leads to an algorithm that produces a stochastic classifier by playing a two-player non-zero-sum game solving for what we call a semi-coarse correlated equilibrium, which in turn corresponds  ...  For example, "Don't recommend a book to a user if it is in a language they haven't purchased before."  ... 
arXiv:1809.04198v1 fatcat:mvzkt5dwvzfwlava2hesyqzb64

Learning to rank for information retrieval

Tie-Yan Liu
2010 Proceeding of the 33rd international ACM SIGIR conference on Research and development in information retrieval - SIGIR '10  
Next, we introduce the benchmark datasets for the research on learning to rank and discuss some practical issues regarding the application of learning to rank, such as click-through log mining and training  ...  In addition, ranking is also pivotal for many other information retrieval applications, such as collaborative filtering, question answering, multimedia retrieval, text summarization, and online advertising  ...  Therefore, learning from click-through logs should also be a online process.  ... 
doi:10.1145/1835449.1835676 dblp:conf/sigir/Liu10 fatcat:ud536unb5ng7pcijqtnlmrxp6y

Introduction to Multi-Armed Bandits [article]

Aleksandrs Slivkins
2022 arXiv   pre-print
Multi-armed bandits a simple but very powerful framework for algorithms that make decisions over time under uncertainty.  ...  Linear contextual bandits have been introduced in Li et al. (2010) , motivated by personalized news recommendations.  ...  ., web search engines, recommendation systems, or online retailers) often compete with one another.  ... 
arXiv:1904.07272v7 fatcat:pptyhyyshrdyhhf7bdonz5dsv4

Web data mining

Kasanda J. Wibonele, Yanqing Zhang, Belur V. Dasarathy
2002 Data Mining and Knowledge Discovery: Theory, Tools, and Technology IV  
., search queries issued by users, and Web pages clicked by visitors.  ...  The concept of duality is widely used in the optimization literature.  ...  For stable classifiers, Bagging may sometime degrade the accuracy.  ... 
doi:10.1117/12.460233 dblp:conf/dmkdttt/WiboneleZ02 fatcat:xc4lmp4xgramljlp45p4ynyvji

A Unified Framework for the Scheduling of Guaranteed Targeted Display Advertising Under Reach and Frequency Requirements

Ali Hojjat, John Turner, Suleyman Cetintas, Jian Yang
2017 Operations Research  
Motivated by recent trends in online advertising and advancements made by online publishers, we consider a new form of contract which allows advertisers to specify the number of unique individuals that  ...  We develop an optimization framework that aims for minimal under-delivery and proper spread of each campaign over its targeted demographics.  ...  Bharadwaj et al. (2012) consider CPM contracts (for which click-through does not play a role) and minimize a weighted objective composed of linear under-delivery and quadratic spreading metrics.  ... 
doi:10.1287/opre.2016.1567 fatcat:dkcmyzwsefejthhrrmybb4rtkm

JSE 32:1 Spring 2018 Whole Issue PDF

Kathleen E. Erickson
2018 Journal of Scientific Exploration  
The authors would like to offer a special thanks to JSE Reviewer Paul Smith and to JSE Associate Editor Roger Nelson for their suggestions and input.  ...  On the backs of three persons there were chalk marks, most frequently in the shape of a question mark. On Siemiradzki's and my shirtcuffs were numerous s-shaped pencil marks.  ...  Lagrangian] of a physical system has a corresponding conservation law." Hamiltonians are generally thought to be equivalent to Lagrangians in this context.  ... 
doaj:fc8d7b487be74877a7bcb333bf9f1e14 fatcat:nreamijiw5ai5low5zl6kp6k2y

Spike-Timing-Dependent Plasticity [chapter]

2017 Encyclopedia of Machine Learning and Data Mining  
, for example, for online shopping or to download music.  ...  Information management encompasses applications such as product recommendation, crime prediction, personalized search, and customer profiling.  ...  We define another operation on case statements max 9E x that is crucial for SDP.  ... 
doi:10.1007/978-1-4899-7687-1_774 fatcat:2jprihjaxfbtpb3ttwuuz3u34y

Sentiment Analysis [chapter]

2017 Encyclopedia of Machine Learning and Data Mining  
, for example, for online shopping or to download music.  ...  Information management encompasses applications such as product recommendation, crime prediction, personalized search, and customer profiling.  ...  We define another operation on case statements max 9E x that is crucial for SDP.  ... 
doi:10.1007/978-1-4899-7687-1_100512 fatcat:ce4yyqo2czftzcx2kbauglh3fu
« Previous Showing results 1 — 15 out of 40 results