1997 search results - page 120 / 400 » On the convergence of Hill's method |
112
click to vote
NIPS
2007
15 years 2 months ago
2007
We present four new reinforcement learning algorithms based on actor-critic and natural-gradient ideas, and provide their convergence proofs. Actor-critic reinforcement learning m...
108
click to vote
CORR
2010
Springer
15 years 1 months ago
2010
Springer
In this paper, we consider the problem of planning and learning in the infinite-horizon discounted-reward Markov decision problems. We propose a novel iterative direct policysearc...
132
click to vote
CVPR
2009
IEEE
15 years 8 months ago
2009
IEEE
Efficient and accurate fitting of Active Appearance Models (AAM) is a key requirement for many applications. The most efficient fitting algorithm today is Inverse Compositiona...
111
Voted
WSC
2000
15 years 2 months ago
2000
Under certain conditions on the integrand, quasi-Monte Carlo methods for estimating integrals (expectations) converge faster asymptotically than Monte Carlo methods. Motivated by ...
141
click to vote
ML
2012
ACM
13 years 8 months ago
2012
ACM
The ratio of two probability densities can be used for solving various machine learning tasks such as covariate shift adaptation (importance sampling), outlier detection (likeliho...