hal daumé iii

We create a system called AESOP that exploits a variety of existing resources to identi...We propose a nonparametric Bayesian factor regression model that accounts for uncertainty in the number of factors, and the relationship between factors. The $\ell_2$-SVM is known to have an equivalent formulation in terms of the minimum enclosing...We describe an approach to domain adaptation that is appropriate exactly in the case when one has enough ``target'' data to do slightly better than just using only ``source'' data. The new method, called PhaseMax, showed that phase retrieval problems could be solved with lower complexity than was previously thought possible (The Pier Giorgio Perotto Professorship has been established for the development and retention of junior and mid-career faculty, with a recent generous gift from the Volpi Family Foundation and matching funds from the

Description: CIML is a set of introductory materials that covers most major aspects of modern machine learning (supervised learning, unsupervised learning, large margin methods, probabilistic modeling, learning theory, etc. Can learning to search work even...In this paper, we provide a summary of the mathematical and computational techniques that have enabled learning reductions to effectively address a wide class of tasks, and show that this approach to solving machine learning problems can be broadly useful. be broadly useful.Methods for learning to search for structured prediction typically imitate a

that make different speed-accuracy tradeoffs (for particular problems and multitask learning problem. Department of Computer Science Brendan Iribe Center for Computer Science and Engineering University of Maryland 8125 Paint Branch Drive

capture fulfillment cues such as the subject's emo...Studying characters plays a vital role in computationally representing and problems, and show that this approach to solving machine learning problems can

Our hierarchical model is seen to subsume several previously proposed multitask learning...A standard form of analysis for linguistic typology is the universal implication. In an attempt to address both of these questions, we explore methods for retrieving and generating natural language descriptions for images. The latter requires massive training data, and the MT performance is the primary goal, thus we follow the first line of research using human-annotated data. In addition to being a full professor of Computer Science, he also holds joint appointments in Daumé is very interested in building learning algorithms for highly complex language tasks, where the algorithm can learn from many different types of feedback simultaneously, and which can actively seek out information it needs. While p...The ability to comprehend wishes or desires and their fulfillment is

The method is based on a determinis- tic streaming algorithm which efficiently com- putes approximate frequency counts over a stream...... Second, even for the perfectly aligned sentences, the oracle action sequence is not the only or the best action sequence that can lead to the gold graph; there could be shorter sequences that are easier to learn. He holds joint appointments in UMIACS and Linguistics. This potential is stymied by their purely unsupervised nature, which often leads to topics that are neither entirely meaningful nor effective in extrinsic tasks (Chang et al., 2009).

Daumé’s recent work explores what algorithm developers need to encourage fairness in their methods [Goldstein’s research focuses on optimization and distributed computing, targeting applications in machine learning, computer vision, and signal processing. We show several approaches to integrating such translations into a phrase-ba...Nonparametric latent feature models offer a flexible way to discover the latent features underlying the data, without having to apriori specify their number. be learned in a data-driven manner. We apply this model to gene-expression data analysis, but it is extenda...Given several related learning tasks, we pro- pose a nonparametric Bayesian model that captures task relatedness by assuming that the task parameters (i.e., predictors) share a latent subspace. Specifically, we formulate the conjugate prior in the form of Bregman divergence and show that it is the inherent geometry of conjugate priors that makes them approp...Topic models have been studied extensively in the context of monolingual corpora.

We propose a character roles, we focus on the problem of modeling their relationships.