Adaptive model selection using empirical complexities

Adaptive Model Selection Using Empirical Complexities

Given n independent replicates of a jointly distributed pair (X; Y ) 2 R R, we wish to select from a xed sequence of model classes F1;F2; : : : a deterministic prediction rule f : R ! R whose risk is small. We investigate the possibility of empirically assessing the complexity of each model class, that is, the actual di culty of the estimation problem within each class. The estimated complexiti...

Adaptive Team Coaching Using Opponent Model Selection

ABSTRACT In multiagent domains with adversarial and ooperative team agents, team agents should be adaptive to the urrent environment and opponent. We introdu e an online method to provide the agents with team plans that a \ oa h" agent generates in response to the spe i opponents. The oa h agent an observe the agents behaviors but it has only periodi ommuni ation with the rest of the team. The ...

Empirical Complexities of Longest Common Subsequence Algorithms

The empirical time complexities of several algorithms for finding the Length of Longest Common Subsequence (LLCS) and the actual Longest Common Subsequence (LCS) of two sequences are examined. These algorithms include a naive recursive algorithm, a recursive method with memoization, dynamic programming, and the Hirschberg LCS algorithm [3]. The empirical time complexities are compared to theore...

Local Rademacher Complexities and Empirical Minimization

Selection of Learning Methods Using an Adaptive Model ofKnowledge

Experiments measuring knowledge utility during the course of learning reveal a common behavior among several learning methods. As the amount of learned knowledge increases, the utility (performance) of the knowledge initially increases, but eventually degrades. A multistrategy learning system based on a model of this common behavior selects a learning method according to the model's prediction ...