A Unified Contextual Bandit Framework for Long- and Short-Term Recommendations

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Authors

We present a unified contextual bandit framework for recommendation problems that is able to capture long- and short-term interests of users. The model is devised in dual space and the derivation is consequentially carried out using Fenchel-Legrende conjugates and thus leverages to a wide range of tasks and settings.We detail two instantiations for regression and classification scenarios and obtain well-known algorithms for these special cases. The resulting general and unified framework allows for quickly adapting contextual bandits to different applications at-hand. The empirical study demonstrates that the proposed long- and short-term framework outperforms both, short-term and long-term models on data. Moreover, a tweak of the combined model proves beneficial in cold start problems.
OriginalspracheEnglisch
TitelMachine Learning and Knowledge Discovery in Databases : European Conference, ECML PKDD 2017 Skopje, Macedonia, September 18 – 22, 2017; Proceedings, Part II
HerausgeberMichelangelo Ceci, Jaakko Hollmen, Ljupco Todorovski, Celine Vens, Saso Dzeroski
Anzahl der Seiten16
Band2
VerlagSpringer
Erscheinungsdatum30.12.2017
Seiten269-284
ISBN (Print)978-3-319-71245-1
ISBN (elektronisch)978-3-319-71246-8
DOIs
PublikationsstatusErschienen - 30.12.2017
VeranstaltungTHE EUROPEAN CONFERENCE ON MACHINE LEARNING & PRINCIPLES AND PRACTICE OF KNOWLEDGE DISCOVERY IN DATABASES 2017 - Skopje, Mazedonien, ehemalige jugoslawische Republik
Dauer: 18.09.201722.09.2017
Konferenznummer: 27
http://ecmlpkdd2017.ijs.si/

DOI

Zuletzt angesehen

Publikationen

  1. Wie beeinflusst die Kameraperspektive die Beurteilung der Unterrichtsqualität?
  2. Microhardness and in vitro corrosion of heat-treated Mg-Y-Ag biodegradable alloy
  3. Lengthscale-dependent modelling of ductile failure in metallic microstructures
  4. Graph-based Approaches for Analyzing Team Interaction on the Example of Soccer
  5. Dual Kalman Filters Analysis for Interior Permanent Magnet Synchronous Motors
  6. Accuracy Improvement by Artificial Neural Networks in Technical Vision System
  7. Friction analyses in twisted and helical profile extrusion of aluminum alloys
  8. Obstacle Coordinates Transformation from TVS Body-Frame to AGV Navigation-Frame
  9. Schülervorstellungen und sozialwissenschaftliche Vorstellungen über Migration
  10. Motion-decoupled internal force control in grasping with visco-elastic contacts
  11. The Lotka-Volterra Model for Competition Controlled by a Sliding Mode Approach
  12. Similar factors underlie tree abundance in forests in native and alien ranges
  13. Do the global stochastic trends drive the real house prices in OECD countries?
  14. Sustainable management of marine fish stocks by means of sliding mode control
  15. Effects of Mn and Zn solutes on grain refinement of commercial pure magnesium
  16. Early Detection of Faillure in Conveyor Chain Systems by Wireless Sensor Node
  17. Impact of soft law regulation by corporate governance codes on firm valuation
  18. Microstructure refinement by a novel friction-based processing on Mg-Zn-Ca alloy
  19. Number theoretical peculiarities in the dimension theory of dynamical systems