360 search results - page 14 / 72 » Combining Learned Discrete and Continuous Action Models |
110
click to vote
NIPS
2008
15 years 1 months ago
2008
We present a discriminative part-based approach for human action recognition from video sequences using motion features. Our model is based on the recently proposed hidden conditi...
ACRI
2004
Springer
15 years 5 months ago
2004
Springer
We present a dynamic, three-dimensional, composite model framework for vertebrate development. Our integrated model combines submodels that address length-scales from subcellular t...
ATAL
2006
Springer
15 years 3 months ago
2006
Springer
We present a method for learning a human understandable, executable model of an agent's behavior using observations of its interaction with the environment. By executable we ...
AR
2007
14 years 11 months ago
2007
—Reinforcement learning is the scheme for unsupervised learning in which robots are expected to acquire behavior skills through self-explorations based on reward signals. There a...
IJRR
2011
14 years 6 months ago
2011
We discuss vision as a sensory modality for systems that effect actions in response to perceptions. While the internal representations informed by vision may be arbitrarily compl...