The Racing Algorithm: Model Selection for Lazy Learners
Artificial Intelligence Review - Special issue on lazy learning
An empirical evaluation of several methods to select the best system
ACM Transactions on Modeling and Computer Simulation (TOMACS)
Simulation Modeling and Analysis
Simulation Modeling and Analysis
Finite-time Analysis of the Multiarmed Bandit Problem
Machine Learning
UAI '04 Proceedings of the 20th conference on Uncertainty in artificial intelligence
Improving the Exploration Strategy in Bandit Algorithms
Learning and Intelligent Optimization
A dynamic programming strategy to balance exploration and exploitation in the bandit problem
Annals of Mathematics and Artificial Intelligence
| Hi-index | 0.00 |
The paper focuses on budgeted model selection, that is the selection between a set of alternative models when the ratio between the number of model assessments and the number of alternatives, though bigger than one, is low. We propose an approach based on the notion of probability of correct selection, a notion borrowed from the domain of Monte Carlo stochastic approximation. The idea is to estimate from data the probability that a greedy selection returns the best alternative and to define a sampling rule which maximises such quantity. Analytical results in the case of two alternatives are extended to a larger number of alternatives by using the Clark's approximation of the maximum of a set of random variables. Preliminary results on synthetic and real model selection tasks show that the technique is competititive with state-of-the-art algorithms, like the bandit UCB.