algorithms to obtain robust predicted click-through rate (CTR) of web objects. The ensemble is acquired by aggregating different pulling policies of bandit algorithms, rather than forcing the agreement of prediction results or learning a unified predictive model. To this end, we employ a meta-bandit paradigm

efficiently even when feedback from consumers is missing or content and preferences evolve over time. Illustrative results highlight the merits of the proposed content aggregation system in a variety of settings. Index Terms—Social multimedia, distributed online learning, content aggregation, multi

This paper analyzes collective decision making when individual preferences evolve through learning. Votes are affected by their anticipated effect on future preferences. The analysis is conducted in a two-arm bandit model with a safe alternative and a risky alternative whose payoff distribution

We study the hiring and retention of heterogeneous workers who learn over time. We show that the problem can be analyzed as an infinite-armed bandit with switching costs and apply results from Employee turnover can similarly affect organizational performance. Workers who turn over (quit

inherits by continuity the properties of the perfect-information benchmark, whereas the other features sizeable heterogeneity of beliefs due to the amplification of private uncertainty through price feedbacks. The two dramatically differ in the impact of shocks at both aggregate and cross-sectional levels

Abstract. Capital goods are embodied knowledge of how to produce. Therefore, capital development is a learning process, through which knowledge gets embodied in new capital goods. Because the necessary knowledge is dispersed among many people who must interact to communicate their particular, often

Key factors for the emergence of collective decision in

We describe Fair Layered Increase/Decrease with Dynamic Layering (FLID-DL), a new multi-rate congestion control algorithm for layered multicast sessions. FLID-DL generalizes the receiver-driven layered congestion (RLC) control protocol introduced by Vicisano, Rizzo, and Crowcroft, ameliorating the problems associated with large IGMP leave latencies and abrupt rate increases. Like RLC, FLID-DL is a scalable, receiver-driven congestion control mechanism in which receivers add layers at sender-initiated synchronization points and leave layers when they experience congestion. FLID-DL congestion control coexists with TCP ows as well as other FLID-DL sessions and supports general rates on the dierent multicast layers. We demonstrate via simulations that our congestion control scheme exhibits better fairness properties and provides better throughput than previous methods.

I herewith certify that all material in this dissertation which is not my own work has been properly acknowledged.

April 29, 2011Für UnsShe had never before seen a rabbit with either a waistcoat-pocket, or a watch to take out of it, and burning with curiosity, she ran across the field after it Lewis CarrollThe following document, written under the supervision of Dr. reviewed by: