Gaia:{Geo-Distributed} machine learning approaching {LAN} speeds
Machine learning (ML) is widely used to derive useful information from large-scale data
(such as user activities, pictures, and videos) generated at increasingly rapid rates, all over …
(such as user activities, pictures, and videos) generated at increasingly rapid rates, all over …
Distributed graphlab: A framework for machine learning in the cloud
While high-level data parallel frameworks, like MapReduce, simplify the design and
implementation of large-scale data processing systems, they do not naturally or efficiently …
implementation of large-scale data processing systems, they do not naturally or efficiently …
[图书][B] Probabilistic graphical models: principles and techniques
D Koller, N Friedman - 2009 - books.google.com
A general framework for constructing and using probabilistic models of complex systems that
would enable a computer to use available information for making decisions. Most tasks …
would enable a computer to use available information for making decisions. Most tasks …
Graphical models, exponential families, and variational inference
MJ Wainwright, MI Jordan - Foundations and Trends® in …, 2008 - nowpublishers.com
The formalism of probabilistic graphical models provides a unifying framework for capturing
complex dependencies among random variables, and building large-scale multivariate …
complex dependencies among random variables, and building large-scale multivariate …
Graphlab: A new framework for parallel machine learning
Designing and implementing efficient, provably correct parallel machine learning (ML)
algorithms is challenging. Existing high-level parallel abstractions like MapReduce are …
algorithms is challenging. Existing high-level parallel abstractions like MapReduce are …
An introduction to conditional random fields
C Sutton, A McCallum - Foundations and Trends® in Machine …, 2012 - nowpublishers.com
Many tasks involve predicting a large number of variables that depend on each other as well
as on other observed variables. Structured prediction methods are essentially a combination …
as on other observed variables. Structured prediction methods are essentially a combination …
Structured learning and prediction in computer vision
S Nowozin, CH Lampert - Foundations and Trends® in …, 2011 - nowpublishers.com
Powerful statistical models that can be learned efficiently from large amounts of data are
currently revolutionizing computer vision. These models possess a rich internal structure …
currently revolutionizing computer vision. These models possess a rich internal structure …
Stereo matching with color-weighted correlation, hierarchical belief propagation, and occlusion handling
In this paper, we formulate a stereo matching algorithm with careful handling of disparity,
discontinuity and occlusion. The algorithm works with a global matching stereo model based …
discontinuity and occlusion. The algorithm works with a global matching stereo model based …
Feature correspondence via graph matching: Models and global optimization
In this paper we present a new approach for establishing correspondences between sparse
image features related by an unknown non-rigid mapping and corrupted by clutter and …
image features related by an unknown non-rigid mapping and corrupted by clutter and …
Dynamic conditional random fields: Factorized probabilistic models for labeling and segmenting sequence data
C Sutton, K Rohanimanesh, A McCallum - Proceedings of the twenty-first …, 2004 - dl.acm.org
In sequence modeling, we often wish to represent complex interaction between labels, such
as when performing multiple, cascaded labeling tasks on the same sequence, or when long …
as when performing multiple, cascaded labeling tasks on the same sequence, or when long …