Fisher Kernel
نویسنده
چکیده
Jaakkola and Haussler (1999a) introduced the Fisher kernel (named in honour of Sir Ronald Fisher), thus creating a generic mechanism for incorporating generative probability models into discriminative classifiers such as SVMs. Jaakkola and Haussler (1999b) introduced a generic class of probabilistic regression models and a parameter estimation technique that can make use of arbitrary kernel functions. Jaakkola, Diekhans and Haussler (1999) applied the Fisher kernel method to detecting remote protein homologies which performed well in classifying protein domains by SCOP superfamily. Jaakkola, Diekhans and Haussler (2000) found that using the Fisher kernel significantly improved on previous methods for the classification of protein domains based on remote homologies. Moreno and Rifkin (2000) used the Fisher kernel method for large scale Web audio classification. Mika, Smola and Schölkopf (2001) presented a fast training algorithm for the kernel Fisher discriminant classifier. It improved upon the state of the art by more than an order of magnitude, thus rendering the kernel Fisher algorithm a viable option also for large datasets. Vinokourov and Girolami (2001) successfully employed the Fisher kernel for document classification. Saunders, Shawe-Taylor and Vinokourov (2003) showed how the string kernel can be thought of as a k-stage Markov process, and as a result interpreted as a Fisher kernel. Tsuda, et al. (2004) analyzed the statistical properties of the Fisher kernel. Nicotra, Micheli and Starita (2004) extended the Fisher kernel to deal with tree structured data. Kersting and Gärtner (2004) extend the Fisher kernel to logical sequences (sequences over an alphabet of logical atoms). Their experiments showed a considerable improvement over results achieved without Fisher kernels for logical sequences. Holub, Welling and Perona (2005) successfully combined generative models with Fisher kernels to realize performance gains on standard object recognition data-sets. The log-likelihood of a data item x with respect to the model m(θ) for a given setting of the parameters θ is defined to be
منابع مشابه
Asymptotic Properties of the Fisher Kernel
This letter analyzes the Fisher kernel from a statistical point of view. The Fisher kernel is a particularly interesting method for constructing a model of the posterior probability that makes intelligent use of unlabeled data (i.e., of the underlying data density). It is important to analyze and ultimately understand the statistical properties of the Fisher kernel. To this end, we first establ...
متن کاملA New Discriminative Kernel From Probabilistic Models
Recently, Jaakkola and Haussler (1999) proposed a method for constructing kernel functions from probabilistic models. Their so-called Fisher kernel has been combined with discriminative classifiers such as support vector machines and applied successfully in, for example, DNA and protein analysis. Whereas the Fisher kernel is calculated from the marginal log-likelihood, we propose the TOP kernel...
متن کاملAn improved training algorithm for kernel Fisher discriminants
We present a fast training algorithm for the kernel Fisher discriminant classifier. It uses a greedy approximation technique and has an empirical scaling behavior which improves upon the state of the art by more than an order of magnitude, thus rendering the kernel Fisher algorithm a viable option also for large datasets.
متن کاملAdaptive Quasiconformal Kernel Fisher Discriminant Analysis via Weighted Maximum Margin Criterion
Kernel Fisher discriminant analysis (KFD) is an effective method to extract nonlinear discriminant features of input data using the kernel trick. However, conventional KFD algorithms endure the kernel selection problem as well as the singular problem. In order to overcome these limitations, a novel nonlinear feature extraction method called adaptive quasiconformal kernel Fisher discriminant ana...
متن کاملThe Fisher Kernel: A Brief Review
The basic idea behind the Fisher kernel method is to train a (generative) hidden Markov model (HMM) on data to derive a Fisher kernel for a (discriminative) support vector machine (SVM). The Fisher kernel gives a ‘natural’ similarity measure that takes into account the underlying probability distribution. If each data item is a (possibly varying length) sequence, each may be used to train a HMM...
متن کامل