Multi-instance (MI) learning is a variant of supervised learning where labeled examples consist of bags (i.e. multi-sets) of feature vectors instead of just a single feature vecto...
We study how to best use crowdsourced relevance judgments learning to rank [1, 7]. We integrate two lines of prior work: unreliable crowd-based binary annotation for binary classi...
The majority of the current information retrieval models weight the query concepts (e.g., terms or phrases) in an unsupervised manner, based solely on the collection statistics. I...
Methods for fusing document lists that were retrieved in response to a query often use retrieval scores (or ranks) of documents in the lists. We present a novel probabilistic fusi...
In this paper, we present a novel, threshold-free robust estimation framework capable of efficiently fitting models to contaminated data. While RANSAC and its many variants have...