A Bayesian Approach to Online Learning

نویسنده

  • Manfred Opper
چکیده

Online learning is discussed from the viewpoint of Bayesian statistical inference. By replacing the true posterior distribution with a simpler parametric distribution, one can define an online algorithm by a repetition of two steps: An update of the approximate posterior, when a new example arrives, and an optimal projection into the parametric family. Choosing this family to be Gaussian, we show that the algorithm achieves asymptotic efficiency. An application to learning in single layer neural networks is given.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A Probabilistic Bayesian Classifier Approach for Breast Cancer Diagnosis and Prognosis

Basically, medical diagnosis problems are the most effective component of treatment policies. Recently, significant advances have been formed in medical diagnosis fields using data mining techniques. Data mining or Knowledge Discovery is searching large databases to discover patterns and evaluate the probability of next occurrences. In this paper, Bayesian Classifier is used as a Non-linear dat...

متن کامل

A Probabilistic Bayesian Classifier Approach for Breast Cancer Diagnosis and Prognosis

Basically, medical diagnosis problems are the most effective component of treatment policies. Recently, significant advances have been formed in medical diagnosis fields using data mining techniques. Data mining or Knowledge Discovery is searching large databases to discover patterns and evaluate the probability of next occurrences. In this paper, Bayesian Classifier is used as a Non-linear dat...

متن کامل

DisTriB: Distributed Trust Management Model Based on Gossip Learning and Bayesian Networks in Collaborative Computing Systems

The interactions among peers in Peer-to-Peer systems as a distributed collaborative system are based on asynchronous and unreliable communications. Trust is an essential and facilitating component in these interactions specially in such uncertain environments. Various attacks are possible due to large-scale nature and openness of these systems that affects the trust. Peers has not enough inform...

متن کامل

DisTriB: Distributed Trust Management Model Based on Gossip Learning and Bayesian Networks in Collaborative Computing Systems

The interactions among peers in Peer-to-Peer systems as a distributed collaborative system are based on asynchronous and unreliable communications. Trust is an essential and facilitating component in these interactions specially in such uncertain environments. Various attacks are possible due to large-scale nature and openness of these systems that affects the trust. Peers has not enough inform...

متن کامل

Investigating university students' views on online learning

Online learning is a concept that has received attention due to new technologies in the field of education; But today, due to the sudden spread of the corona virus, online learning has become common, so that most of the higher education institutions organize online learning courses. However, for many students, especially new undergraduate students who are used to the traditional learning enviro...

متن کامل

Bayesian online learning in

In a Bayesian approach to online learning a simple approximate parametric form for posterior is updated in each online learning step. Usually in online learning only an estimate of the solution is updated. The Bayesian online approach is applied to two simple learning scenarios, learning a perceptron rule with respectively a spherical and a binary weight prior. In the rst case we rederive the r...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2006