Results 1 
5 of
5
Semirational Models of Conditioning: The Case of Trial Order
, 2007
"... Bayesian treatments of animal conditioning start from a generative model that specifies precisely a set of assumptions about the structure of the learning task. Optimal rules for learning are direct mathematical consequences of these assumptions. In terms of Marr’s (1982) levels of analyses, the mai ..."
Abstract

Cited by 4 (2 self)
 Add to MetaCart
Bayesian treatments of animal conditioning start from a generative model that specifies precisely a set of assumptions about the structure of the learning task. Optimal rules for learning are direct mathematical consequences of these assumptions. In terms of Marr’s (1982) levels of analyses, the main task at the computational level
Bayesian clustering of huge friendship networks
, 2007
"... Because of the recent growth in popularity of social websites, such as MySpace, Facebook and Last.fm, there is an increasing interest in ways to analyze extremely large friendship networks with even millions of nodes. These huge networks provide a practical test ground for new network algorithms. Th ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
Because of the recent growth in popularity of social websites, such as MySpace, Facebook and Last.fm, there is an increasing interest in ways to analyze extremely large friendship networks with even millions of nodes. These huge networks provide a practical test ground for new network algorithms. The network analysis methods can also be applied to other networks than social networks, such as interactions between proteins and links between web pages. Social networks have typically structure: there are dense groups of nodes and some nodes have disproportionately many connections. The structure emerges, because friendships are not formed randomly. Instead, people tend to become friends with those who are similar to themselves. This can be called homophily. There are also other factors that guide the formation of friendships, such as geographical location and membership in common activities. The M0 algorithm finds clustering structure in networks with homophily by Bayesian statistical inference. The algorithm is based on a generative model for creating the edges
Efficient Boundary Tracking Through Sampling
"... The proposed algorithm for image segmentation is inspired by an algorithm for autonomous environmental boundary tracking. The algorithm relies on a tracker that traverses a boundary between regions in a sinusoidallike path. Boundary tracking is done by efficiently sampling points, resulting in a si ..."
Abstract
 Add to MetaCart
The proposed algorithm for image segmentation is inspired by an algorithm for autonomous environmental boundary tracking. The algorithm relies on a tracker that traverses a boundary between regions in a sinusoidallike path. Boundary tracking is done by efficiently sampling points, resulting in a significant savings in computation time. Page’s cumulative sum (CUSUM) procedure and other methods are adapted to handle a high level of noise. Applications to large data sets such as hyperspectral are of particular interest. Irregularly shaped boundaries such as fractals are also treated at very fine detail.
doi:10.1093/imrn/rnn000 Efficient Boundary Tracking Through Sampling
"... The proposed algorithm for image segmentation is inspired by an algorithm for autonomous environmental boundary tracking. The algorithm relies on a tracker that traverses a boundary between regions in a sinusoidallike path. Boundary tracking is done by efficiently sampling points, resulting in a si ..."
Abstract
 Add to MetaCart
The proposed algorithm for image segmentation is inspired by an algorithm for autonomous environmental boundary tracking. The algorithm relies on a tracker that traverses a boundary between regions in a sinusoidallike path. Boundary tracking is done by efficiently sampling points, resulting in a significant savings in computation time over many other segmentation methods. For noisy images, the traversed path is modeled as a changepoint detection problem between two states. Changepoint detection algorithms such as Page’s cumulative sum procedure are adapted in conjunction with other methods to handle a high level of noise. A modification for the multipleregion case is also presented as a hybrid of a topologydetecting segmentation algorithm and boundary tracking. Applications to high resolution images and large data sets such as hyperspectral are of particular interest. Irregularly shaped boundaries such as fractals are also treated at very fine detail along with accompanying fractal dimension calculations, which follow naturally from the boundary tracking data. 1