14,426 Hits in 3.0 sec

Exploration-exploitation tradeoff in interactive relevance feedback

Maryam Karimzadehgan, ChengXiang Zhai
2010 Proceedings of the 19th ACM international conference on Information and knowledge management - CIKM '10  
Optimizing such an exploration-exploitation tradeoff is key to the optimization of the overall utility of relevance feedback to a user in the entire session of relevance feedback.  ...  Experiment results show that the proposed learning approach can effectively optimize the exploration-exploitation tradeoff and outperforms the traditional relevance feedback approach which only does exploitation  ...  This work is a first step in addressing the new problem of optimizing the exploration-exploitation tradeoff in interactive feedback. There are many interesting directions to further explore.  ... 
doi:10.1145/1871437.1871631 dblp:conf/cikm/KarimzadehganZ10 fatcat:et67cgnmfzfw3ccszpcepiy64u

Unpacking the exploration–exploitation tradeoff: A synthesis of human and animal literatures

Katja Mehlhorn, Ben R. Newell, Peter M. Todd, Michael D. Lee, Kate Morgan, Victoria A. Braithwaite, Daniel Hausmann, Klaus Fiedler, Cleotilde Gonzalez
2015 Decision  
The balance required in these situations is commonly referred to as the exploration-exploitation tradeoff.  ...  Here, we integrate findings from these and other often-isolated literatures in order to gain a better understanding of the possible tradeoffs between exploration and exploitation, and we propose new theoretical  ...  feedback information and rewards in a given trial.  ... 
doi:10.1037/dec0000033 fatcat:iyzharxp2jf2ti5hyczt6hjnuq

Supporting exploratory search tasks with interactive user modeling

Tuukka Ruotsalo, Kumaripaba Athukorala, Dorota Głowacka, Ksenia Konyushkova, Antti Oulasvirta, Samuli Kaipiainen, Samuel Kaski, Giulio Jacucci
2013 Proceedings of the American Society for Information Science and Technology  
Contrary to traditional interactions, such as query based search, query suggestions, or relevance feedback, interactive user modeling allows a user to perceive the state of the user model at all times  ...  The results show that interactive user modeling can help users to more effectively find relevant, novel and diverse information without compromises in task execution time.  ...  The data used in the experiments is derived from the Web of Science prepared by THOMSON REUTERS, Inc., Philadelphia, Pennsylvania, USA: Copyright THOMSON REUTERS, 2011.  ... 
doi:10.1002/meet.14505001040 fatcat:tjv5wa3nzngrfilefl5ccwzi2m

Improving the Initial Image Retrieval Set by Inter-Query Learning with One-Class SVMs [chapter]

Iker Gondra, Douglas R. Heisterkamp, Jing Peng
2003 Intelligent Systems Design and Applications  
Relevance Feedback attempts to reduce the semantic gap between a user's perception of similarity and a feature-based representation of an image by asking the user to provide feedback regarding the relevance  ...  In this paper, we focus on the possibility of incorporating prior experience (obtained from the historical interaction of users with the system) to improve the retrieval performance on future queries.  ...  Relevance feedback attempts to overcome these problems by gathering semantic information from user interaction.  ... 
doi:10.1007/978-3-540-44999-7_38 fatcat:3o4qrzxv4zgo3mxlvuqdzyrw6e

Interactive Intent Modeling for Exploratory Search

Tuukka Ruotsalo, Jaakko Peltonen, Manuel J. A. Eugster, Dorota Głowacka, Patrik Floréen, Petri Myllymäki, Giulio Jacucci, Samuel Kaski
2018 ACM Transactions on Information Systems  
Consequently, users can discover and explore information relevant to their tasks rather than optimizing results to be maximally relevant for an individual query, which may be suboptimal in the first place  ...  The needs of the user may evolve throughout the search session, and the user may need assistance in directing the search to explore initially unpredictable but highly relevant information.  ...  We thank all researchers participating in HWFA, Multivire, D2I, and Re: Know for their help and assistance.  ... 
doi:10.1145/3231593 fatcat:ss7vw5p7yfggvkti3rpaia4jrm

Visual analytics clarify the scalability and effectiveness of massively parallel many-objective optimization: A groundwater monitoring design example

Patrick M. Reed, Joshua B. Kollat
2013 Advances in Water Resources  
Unlike prior approaches to water resources planning and management, massively parallel many-objective visual analytics benefits in its parallel scaling and range of tradeoffs explored with increases in  ...  Our study contributes a promising computational platform for providing rapid and evolving feedbacks between science, engineering, and decision makers for problem complexities relevant in a nonstationary  ... 
doi:10.1016/j.advwatres.2013.01.011 fatcat:fi2imq573rc5xcec4rvgsfipdq

Explore-exploit in top-N recommender systems via Gaussian processes

Hastagiri P. Vanchinathan, Isidor Nikolic, Fabio De Bona, Andreas Krause
2014 Proceedings of the 8th ACM Conference on Recommender systems - RecSys '14  
We propose the CGPRANK algorithm, which exploits prior information specified in terms of a Gaussian process kernel function, which allows to share feedback in three ways: Between positions in a list, between  ...  In our experiments, our CGPRANK approach significantly outperforms state-of-the-art multi-armed bandit and learning-to-rank methods, with an 18% increase in clicks.  ...  CGPRANK navigates the exploration-exploitation tradeoff by predicting performance of as yet unexplored lists using nonparametric Gaussian process models, whose regularity is captured in the kernel function  ... 
doi:10.1145/2645710.2645733 dblp:conf/recsys/VanchinathanNBK14 fatcat:fcg3jbnvhvbcvga5l6jg4qxjeq

Advertising keyword generation using active learning

Hao Wu, Guang Qiu, Xiaofei He, Yuan Shi, Mingcheng Qu, Jing Shen, Jiajun Bu, Chun Chen
2009 Proceedings of the 18th international conference on World wide web - WWW '09  
This paper proposes an efficient relevance feedback based interactive model for keyword generation in sponsored search advertising.  ...  We formulate the ranking of relevant terms as a supervised learning problem and suggest new terms for the seed by leveraging user relevance feedback information.  ...  Recent related work tends to exploit semantic relation- * Supported by the National Key Technology R&D Program In this paper, we propose an interactive model to explore relevance feedback for keyword generation  ... 
doi:10.1145/1526709.1526873 dblp:conf/www/WuQHSQSBC09 fatcat:gorfj5qpcbex7dyyxo7kjikhsa

Iterative relevance feedback with adaptive exploration/exploitation trade-off

Nicolae Suditu, François Fleuret
2012 Proceedings of the 21st ACM international conference on Information and knowledge management - CIKM '12  
Here, in contrast with other types of retrieval systems, these two regimes are of great importance since the search initialization is hardly optimal (i.e. the page-zero problem) and the relevance feedback  ...  Starting from a query-free approach meant to solve the page-zero problem, we propose an adaptive exploration/exploitation trade-off that transforms the original framework into a versatile retrieval framework  ...  Although the "sampling" algorithm does not cover the entire collection anymore, the system continuously estimates the exploration/exploitation tradeoff that suits the user.  ... 
doi:10.1145/2396761.2398435 dblp:conf/cikm/SudituF12 fatcat:pfbx76t6ijdmvccb4ediorgycy

Exploitation and exploration in a performance based contextual advertising system

Wei Li, Xuerui Wang, Ruofei Zhang, Ying Cui, Jianchang Mao, Rong Jin
2010 Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '10  
The exploitation and exploration (EE) tradeoff has been extensively studied in the reinforcement learning community, however, not been paid much attention in online advertising until recently.  ...  more about the unknown to improve its knowledge (exploration), since the latter might increase its revenue in the future.  ...  RELATED WORK The exploitation and exploration tradeoff was first formally studied in reinforcement learning in 1980's, and later flourished in other fields of machine learning [12, 13] .  ... 
doi:10.1145/1835804.1835811 dblp:conf/kdd/LiWZCMJ10 fatcat:7rupiwrdovcpjepj55naiedeje

Balance Within and Across Domains: The Performance Implications of Exploration and Exploitation in Alliances

Dovev Lavie, Jingoo Kang, Lori Rosenkopf
2011 Organization science (Providence, R.I.)  
We posit that firms can overcome such impediments and enhance their performance if they explore in one domain while exploiting in another.  ...  We posit that firms can overcome such impediments and enhance their performance if they explore in one domain while exploiting in another.  ...  In particular, a firm may face resource allocation tradeoffs when balancing exploration and exploitation within the function domain.  ... 
doi:10.1287/orsc.1100.0596 fatcat:e4k4wzuoqvanrg7xrawqszpksa

Interactive Symptom Elicitation for Diagnostic Information Retrieval

Tuukka Ruotsalo, Antti Lipsanen
2018 The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval - SIGIR '18  
Medical information retrieval suffers from a dual problem: users struggle in describing what they are experiencing from a medical perspective and the search engine is struggling in retrieving the information  ...  We demonstrate interactive symptom elicitation for diagnostic information retrieval.  ...  The algorithm controls the exploration/exploitation tradeoff of the estimates: it predicts expected symptom relevances (exploiting symptoms that have the best fit given the user feedback) and corresponding  ... 
doi:10.1145/3209978.3210172 dblp:conf/sigir/RuotsaloL18 fatcat:dwdfbvz2wrawhn6k5xtn4bsosi

Interactive Exploration for Image Retrieval

Matthieu Cord, Sylvie Philipp-Foliguet, Philippe-Henri Gosselin, Jérôme Fournier
2005 EURASIP Journal on Advances in Signal Processing  
First, a controlled exploration method of the database is presented. Second, a relevance feedback method based on statistical learning is proposed.  ...  Using these powerful indexes, an original interactive retrieval strategy is introduced. The process is based on two steps for handling the retrieval of very large image categories.  ...  They are used after the exploration and compared to the relevance feedback technique without exploration.  ... 
doi:10.1155/asp.2005.2173 fatcat:hb47phaysve3bb6evqfxlzbvnu


Yanlei Diao, Kyriaki Dimitriadou, Zhan Li, Wenzhao Liu, Olga Papaemmanouil, Kemi Peng, Liping Peng
2015 Proceedings of the VLDB Endowment  
AIDE steers the user towards interesting data areas based on her relevance feedback on database samples, aiming to achieve the goal of identifying all database objects that match the user interest with  ...  In our demonstration, conference attendees will see AIDE in action for a variety of exploration tasks on real-world datasets.  ...  ACKNOWLEDGMENTS This work was funded in part by NSF under grants IIS-1253196 and IIS-1218524.  ... 
doi:10.14778/2824032.2824112 fatcat:qyq4ml3opfhe7o53rwrhwfutjq

Directing exploratory search with interactive intent modeling

Tuukka Ruotsalo, Giulio Jacucci, Samuel Kaski, Jaakko Peltonen, Manuel Eugster, Dorota Głowacka, Ksenia Konyushkova, Kumaripaba Athukorala, Ilkka Kosunen, Aki Reijonen, Petri Myllymäki
2013 Proceedings of the 22nd ACM international conference on Conference on information & knowledge management - CIKM '13  
We systematically evaluated the effect of the interactive intent modeling in a mixed-method task-based information seeking setting with 30 users, where we compared two interface variants for interactive  ...  We introduce interactive intent modeling, where the user directs exploratory search by providing feedback for estimates of search intents.  ...  To deal with the exploration-exploitation tradeoff we select keywords not with the highest relevance score, but with the largest upper confidence bound for the score.  ... 
doi:10.1145/2505515.2505644 dblp:conf/cikm/RuotsaloPEGKAKRMJK13 fatcat:yd33vatex5copmg7whoq7cix4u
« Previous Showing results 1 — 15 out of 14,426 results