搜索结果: 1-7 共查到“统计学 Online Learning”相关记录7条 . 查询时间(0.14 秒)
We introduce online learning algorithms which are independent of feature scales, proving regret bounds dependent on the ratio of scales existent in the data rather than the absolute scale. This has se...
Online Learning in a Contract Selection Problem
Online Learning Contract Selection Problem
2013/6/14
In an online contract selection problem there is a seller which offers a set of contracts to sequentially arriving buyers whose types are drawn from an unknown distribution. If there exists a profitab...
On the Generalization Ability of Online Learning Algorithms for Pairwise Loss Functions
Generalization Ability Online Learning Algorithms Pairwise Loss Functions
2013/6/14
In this paper, we study the generalization properties of online learning based stochastic methods for supervised learning problems where the loss function is dependent on more than one training sample...
Online Learning in Markov Decision Processes with Adversarially Chosen Transition Probability Distributions
Online Learning Markov Decision Processes Adversarially Chosen Transition Probability Distributions
2013/5/2
We study the problem of learning Markov decision processes with finite state and action spaces when the transition probability distributions and loss functions are chosen adversarially and are allowed...
Second-Order Non-Stationary Online Learning for Regression
Second-Order Non-Stationary Online Learning for Regression
2013/4/28
The goal of a learner, in standard online learning, is to have the cumulative loss not much larger compared with the best-performing function from some fixed class. Numerous algorithms were shown to h...
We present methods for online linear optimization that take advantage of benign (as opposed to worst-case) sequences. Speci cally if the sequence encountered by the learner is described well by a know...
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.