搜索结果: 46-60 共查到“统计学 learning”相关记录102条 . 查询时间(0.14 秒)
Distance Metric Learning for Kernel Machines
metric learning distance learning support vector machines semi-denite programming Mahalanobis distance
2012/9/17
Recent work in metric learning has signicantly improved the state-of-the-art ink-nearest neighbor classication. Support vector machines (SVM), particularly with RBF kernels, are amongst the most pop...
Detecting Events and Patterns in Large-Scale User Generated Textual Streams with Statistical Learning Methods
Detecting Events Patterns Large-Scale User Generated Textual Streams Statistical Learning Methods
2012/9/18
A vast amount of textual web streams is influenced by events or phenomena emerging in the real world. The social web forms an excellent modern paradigm, where unstructured user generated content is pu...
One Permutation Hashing for Efficient Search and Learning
Permutation Hashing Efficient Search Learning
2012/9/18
Minwise hashing is a standard procedure in the context of search, for efficiently estimating set similari-ties in massive binary data such as text. Recently, the method ofb-bit minwise hashing has bee...
Learning Theory Approach to Minimum Error Entropy Criterion
minimum error entropy learning theory Renyi’s entropy empirical risk minimization approximation error
2012/9/18
We consider the minimum error entropy (MEE) criterion and anempirical risk minimization learning algorithm in a regression setting. Alearning theory approach is presented for this MEE algorithm and ex...
Parameter and Structure Learning in Nested Markov Models
Parameter Structure Learning in Nested Markov Models
2012/9/19
The constraints arising from DAG mod-els with latent variables can be naturally represented by means of acyclic directed mixed graphs (ADMGs). Such graphs contain directed (!) and bidirected ($) arrow...
Surrogate Losses in Passive and Active Learning
active learning sequential design selective sampling statistical learning theory surrogate loss functions classification
2012/9/19
Active learning is a type of sequential design for supervised machine learning, in which the learning algorithm sequentially requests the labels of selected instances from a large pool of unlabeled da...
We consider a basic problem in unsupervised learning: learning an unknown \emph{Poisson Binomial Distribution} over $\{0,1,...,n\}$. A Poisson Binomial Distribution (PBD) is a sum $X = X_1 + ... + X_n...
A $k$-modal probability distribution over the domain $\{1,...,n\}$ is one whose histogram has at most $k$ "peaks" and "valleys." Such distributions are natural generalizations of monotone ($k=0$) and ...
Provably Safe and Robust Learning-Based Model Predictive Control
Safe and Robust Learning-Based Model Predictive Control
2011/7/19
Controller design for systems typically faces a trade-off between robustness and performance, and the reliability of linear controllers has caused many control practitioners to focus on the former. Ho...
Multi-Instance Learning with Any Hypothesis Class
Multiple-instance learning learning theory sample complexity PAC learning
2011/7/19
In the supervised learning setting termed Multiple-Instance Learning (MIL), the examples are bags of instances, and the bag label is a function of the labels of its instances. Typically, this function...
Spectral Methods for Learning Multivariate Latent Tree Structure
Spectral Methods Learning Multivariate Latent Tree Structure
2011/7/19
This work considers the problem of learning the structure of a broad class of multivariate latent variable tree models, which include a variety of continuous and discrete models (including the widely ...
Spectral Methods for Learning Multivariate Latent Tree Structure
Multivariate Latent Spectral Methods
2011/7/19
This work considers the problem of learning the structure of a broad class of multivariate latent variable tree models, which include a variety of continuous and discrete models (including the widely ...
Distinct counting with a self-learning bitmap
Distinct counting sampling streaming data bitmap
2011/7/19
Counting the number of distinct elements (cardinality) in a dataset is a fundamental problem in database management. In recent years, due to many of its modern applications, there has been significant...
Learning with the Weighted Trace-norm under Arbitrary Sampling Distributions
Learning Weighted Trace-norm Arbitrary Sampling Distributions
2011/7/7
We provide rigorous guarantees on learning with the weighted trace-norm under arbitrary sampling distributions.
Efficient Online Learning via Randomized Rounding
Efficient Online Learning Randomized Rounding
2011/7/6
Most online algorithms used in machine learning today are based on variants of mirror descent or follow-the-leader.