113
Voted
ICML
1998
IEEE
16 years 3 months ago
1998
IEEE
tion Learning about Temporally Abstract Actions Richard S. Sutton Department of Computer Science University of Massachusetts Amherst, MA 01003-4610 rich@cs.umass.edu Doina Precup D...
118
Voted
ICML
1998
IEEE
16 years 3 months ago
1998
IEEE
Stochastic topological models, and hidden Markov models in particular, are a useful tool for robotic navigation and planning. In previous work we have shown how weak odometric dat...
135
Voted
ICML
1998
IEEE
16 years 3 months ago
1998
IEEE
Production scheduling, the problem of sequentially con guring a factory to meet forecasted demands, is a critical problem throughout the manufacturing industry. The requirement of...
93
Voted
ICML
1998
IEEE
16 years 3 months ago
1998
IEEE
133
Voted
ICML
1998
IEEE
16 years 3 months ago
1998
IEEE
This paper introduces the RL-TOPs architecture for robot learning, a hybrid system combining teleo-reactive planning and reinforcement learning techniques. The aim of this system ...
116
Voted
ICML
1998
IEEE
16 years 3 months ago
1998
IEEE
We analyze critically the use of classi cation accuracy to compare classi ers on natural data sets, providing a thorough investigation using ROC analysis, standard machine learnin...
97
Voted
ICML
1998
IEEE
16 years 3 months ago
1998
IEEE
Three factors are related in analyses of performance curves such as learning curves: the amount of training, the learning algorithm, and performance. Often we want to know whether...
91
Voted
ICML
1998
IEEE
16 years 3 months ago
1998
IEEE
121
Voted
ICML
1998
IEEE
16 years 3 months ago
1998
IEEE
This paper introduces a new algorithm, Q2, foroptimizingthe expected output ofamultiinput noisy continuous function. Q2 is designed to need only a few experiments, it avoids stron...
99
Voted
ICML
1998
IEEE
16 years 3 months ago
1998
IEEE