This paper explores two classes of model adaptation methods for Web search ranking: Model Interpolation and error-driven learning approaches based on a boosting algorithm. The res...
Jianfeng Gao, Qiang Wu, Chris Burges, Krysta Marie...
—In lattice-coded multiple-input multiple-output (MIMO) systems, optimal decoding amounts to solving the closest vector problem (CVP). Embedding is a powerful technique for the a...
Cong Ling, Shuiyin Liu, Laura Luzzi, Damien Stehl&...
Most effective particular object and image retrieval approaches are based on the bag-of-words (BoW) model. All state-of-the-art retrieval results have been achieved by methods tha...
Ondrej Chum, Andrej Mikulik, Michal Perdoch, Jiri ...
—To overcome the limitations of specific positioning techniques for mobile wireless nodes and achieve a high accuracy, the fusion of heterogeneous sensor information is an appea...
—In this paper we present a methodology for the design of congestion control protocols for background data transfers that have a minimal delay impact on short TCP transfers and c...