Online to Batch Conversion
نویسنده
چکیده
Traditional supervised learning is formulated as learning from a given data set while being able to generalize to unseen data. It is usually assumed that both the given and unseen data are drawn iid from the same (unknown) distribution. In online learning, we make no assumption about the source of data. One simply observes a stream of data coming from some arbitrary source one by one. At every step, the online learning algorithm tries to make a correct prediction for the next data point. After making the prediction, it observes the true label and updates its hypothesis in light of the new evidence. The goal is to minimize regret:
منابع مشابه
Data-Driven Online to Batch Conversions
Online learning algorithms are typically fast, memory efficient, and simple to implement. However, many common learning problems fit more naturally in the batch learning setting. The power of online learning algorithms can be exploited in batch settings by using online-to-batch conversions, techniques which build a new batch algorithm from an existing online algorithm. We first give a unified o...
متن کاملFrom Batch to Transductive Online Learning
It is well-known that everything that is learnable in the difficult online setting, where an arbitrary sequences of examples must be labeled one at a time, is also learnable in the batch setting, where examples are drawn independently from a distribution. We show a result in the opposite direction. We give an efficient conversion algorithm from batch to online that is transductive: it uses futu...
متن کاملFrom Online to Batch Learning with Cutoff-Averaging
We present cutoff averaging, a technique for converting any conservative online learning algorithm into a batch learning algorithm. Most online-to-batch conversion techniques work well with certain types of online learning algorithms and not with others, whereas cutoff averaging explicitly tries to adapt to the characteristics of the online algorithm being converted. An attractive property of o...
متن کاملFrom Online to Batch Learning with Cutoff-Averaging
We present cutoff averaging, a technique for converting any conservative online learning algorithm into a batch learning algorithm. Most online-to-batch conversion techniques work well with certain types of online learning algorithms and not with others, whereas cutoff averaging explicitly tries to adapt to the characteristics of the online algorithm being converted. An attractive property of o...
متن کاملStructural Online Learning
We study the problem of learning ensembles in the online setting, when the hypotheses are selected out of a base family that may be a union of possibly very complex sub-families. We prove new theoretical guarantees for the online learning of such ensembles in terms of the sequential Rademacher complexities of these sub-families. We also describe an algorithm that benefits from such guarantees. ...
متن کاملFrom exp-concavity to variance control: High probability O(1/n) rates and high probability online-to-batch conversion
We present an algorithm for the statistical learning setting with a bounded exp-concave loss in d dimensions that obtains excess risk O(d/n) with high probability: the dependence on the confidence parameter δ is polylogarithmic in 1/δ. The core technique is to boost the confidence of recent O(d/n) bounds, without sacrificing the rate, by leveraging a Bernstein-type condition which holds due to ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2011