Package org.drip.learning.rxtor1

Statistical Learning Empirical Loss Penalizer
Author:
Lakshmi Krishnamurthy
  • Interface Summary
    Interface Description
    EmpiricalLearningMetricEstimator
    EmpiricalLearningMetricEstimator is the Estimator of the Empirical Loss and Risk, as well as the corresponding Covering Numbers.
  • Class Summary
    Class Description
    ApproximateLipschitzLossLearner
    ApproximateLipschitzLossLearner implements the Learner Class that holds the Space of Normed Rd To Normed R1 Learning Functions for the Family of Loss Functions that are "approximately" Lipschitz, i.e., loss (ep) - loss (ep') Less Than max (C * |ep-ep'|, C')

    The References are:

    Alon, N., S.
    EmpiricalPenaltySupremum
    EmpiricalPenaltySupremum holds the Learning Function that corresponds to the Empirical Supremum, as well as the corresponding Supremum Value.
    EmpiricalPenaltySupremumEstimator
    EmpiricalPenaltySupremumEstimator contains the Implementation of the Empirical Penalty Supremum Estimator dependent on Multivariate Random Variables where the Multivariate Function is a Linear Combination of Bounded Univariate Functions acting on each Random Variate.
    EmpiricalPenaltySupremumMetrics
    EmpiricalPenaltySupremumMetrics computes Efron-Stein Metrics for the Penalty Supremum Rx To R1 Functions.
    GeneralizedLearner
    GeneralizedLearner implements the Learner Class that holds the Space of Normed Rx To Normed R1 Learning Functions along with their Custom Empirical Loss.
    L1LossLearner
    L1LossLearner implements the Learner Class that holds the Space of Normed Rx To Normed R1 Learning Functions that employs L1 Empirical Loss Routine.
    LipschitzLossLearner
    LipschitzLossLearner implements the Learner Class that holds the Space of Normed R1 To Normed R1 Learning Functions for the Family of Loss Functions that are Lipschitz, i.e., loss (ep) - loss (ep') Less Than C * |ep-ep'|

    The References are:

    Alon, N., S.
    LpLossLearner
    LpLossLearner implements the Learner Class that holds the Space of Normed Rx To Normed R1 Learning Functions for the Family of Loss Functions that are Polynomial, i.e., loss (eta) = (eta ^ p) / p, for p greater than 1.