Name
Abstract | ||
|---|---|---|
We propose an alternative parameterization of Logistic Regression (LR) for the categorical data, multi-class setting. LR optimizes the conditional log-likelihood over the training data and is based on an iterative optimization procedure to tune this objective function. The optimization procedure employed may be sensitive to scale and hence an effective pre-conditioning method is recommended. Many problems in machine learning involve arbitrary scales or categorical data (where simple standardization of features is not applicable). The problem can be alleviated by using optimization routines that are invariant to scale such as (second-order) Newton methods. However, computing and inverting the Hessian is a costly procedure and not feasible for big data. Thus one must often rely on first-order methods such as gradient descent (GD), stochastic gradient descent (SGD) or approximate second-order such as quasi-Newton (QN) routines, which are not invariant to scale. This paper proposes a simple yet effective pre-conditioner for speeding-up LR based on naive Bayes conditional probability estimates. The idea is to scale each attribute by the log of the conditional probability of that attribute given the class. This formulation substantially speeds-up LR's convergence. It also provides a weighted naive Bayes formulation which yields an effective framework for hybrid generative-discriminative classification. |
Year | DOI | Venue |
|---|---|---|
2014 | 10.1109/ICDM.2014.53 | Data Mining |
Keywords | Field | DocType |
Bayes methods,Newton method,convergence,learning (artificial intelligence),optimisation,pattern classification,regression analysis,LR convergence,categorical data,conditional log-likelihood,hybrid generative-discriminative classification,iterative optimization,logistic regression,machine learning,multiclass setting,naive Bayes conditional probability estimates,naive-Bayes inspired effective preconditioner,optimization routines,parameterisation,preconditioning method,second-order Newton methods,weighted naive Bayes formulation,classification,discriminative-generative learning,logistic regression,pre-conditioning,stochastic gradient descent,weighted naive Bayes | Convergence (routing),Data mining,Gradient descent,Stochastic gradient descent,Conditional probability,Naive Bayes classifier,Categorical variable,Computer science,Hessian matrix,Artificial intelligence,Invariant (mathematics),Machine learning | Conference |
ISSN | ISBN | Citations |
1550-4786 | 978-1-4799-4303-6 | 5 |
PageRank | References | Authors |
0.47 | 3 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Nayyar Abbas Zaidi | 1 | 91 | 9.88 |
| Mark Carman | 2 | 563 | 49.18 |
| Jesus Cerquides | 3 | 280 | 27.74 |
| Geoffrey I. Webb | 4 | 3130 | 234.10 |