ICML
2009
IEEE
14 years 5 months ago
2009
IEEE
We consider the problem of estimating the policy gradient in Partially Observable Markov Decision Processes (POMDPs) with a special class of policies that are based on Predictive ...
ICML
2009
IEEE
14 years 5 months ago
2009
IEEE
Ranking a set of retrieved documents according to their relevance to a query is a popular problem in information retrieval. Methods that learn ranking functions are difficult to o...
ICML
2009
IEEE
14 years 5 months ago
2009
IEEE
Previous studies of Non-Parametric Kernel (NPK) learning usually reduce to solving some Semi-Definite Programming (SDP) problem by a standard SDP solver. However, time complexity ...
ICML
2009
IEEE
14 years 5 months ago
2009
IEEE
In this paper we introduce the first algorithms for efficiently learning a simulation policy for Monte-Carlo search. Our main idea is to optimise the balance of a simulation polic...
ICML
2009
IEEE
14 years 5 months ago
2009
IEEE
Approximate Linear Programming (ALP) is a reinforcement learning technique with nice theoretical properties, but it often performs poorly in practice. We identify some reasons for...
ICML
2009
IEEE
14 years 5 months ago
2009
IEEE
A key aspect of semantic image segmentation is to integrate local and global features for the prediction of local segment labels. We present an approach to multi-class segmentatio...
ICML
2009
IEEE
14 years 5 months ago
2009
IEEE
We propose an unbounded-depth, hierarchical, Bayesian nonparametric model for discrete sequence data. This model can be estimated from a single training sequence, yet shares stati...
ICML
2009
IEEE
14 years 5 months ago
2009
IEEE
This paper describes an unsupervised learning technique for modeling human locomotion styles, such as distinct related activities (e.g. running and striding) or variations of the ...
ICML
2009
IEEE
14 years 5 months ago
2009
IEEE
In this paper, we investigate a simple, mistakedriven learning algorithm for discriminative training of continuous density hidden Markov models (CD-HMMs). Most CD-HMMs for automat...
ICML
2009
IEEE
14 years 5 months ago
2009
IEEE
In many real-world domains, undirected graphical models such as Markov random fields provide a more natural representation of the dependency structure than directed graphical mode...