Skip to main content

Module types

Module types 

Source
Expand description

Auto-generated module

🤖 Generated with SplitRS

Structs§

AdaBoost
AdaBoost: adaptive boosting with exponential loss.
BiasVarianceTradeoff
Bias-variance tradeoff decomposition: MSE = Bias² + Variance + Noise.
CausalBackdoor
Backdoor adjustment formula for causal inference.
CrossValidation
Cross-validation scheme.
DoubleRademacher
Double Rademacher (two-sided) bound.
ELBO
Evidence Lower Bound (ELBO) for variational inference.
EarlyStoppingReg
Early stopping regularization — implicit regularization by iteration count.
ExponentialWeightsAlgorithm
Exponential Weights Algorithm (Hedge / EWA) for online learning.
FeatureMap
Feature map: maps inputs to a (truncated) explicit feature space.
FisherInformation
Fisher information I(θ) = E[(∂/∂θ log p(x;θ))²].
GaussianComplexity
Gaussian complexity (Gaussian analog of Rademacher).
GaussianProcess
Gaussian process regression model.
GradientBoosting
Ensemble method: gradient boosting.
GrowthFunction
Growth function Π_H(m) for a hypothesis class.
KLDivergence
KL divergence D_KL(P‖Q) = Σ P(x) log(P(x)/Q(x)).
KernelFunction
A kernel function k: ℝ^d × ℝ^d → ℝ.
KernelMatrix
Kernel (Gram) matrix K_{ij} = k(x_i, x_j).
KernelSVM
Kernel SVM dual representation.
KernelSVMTrainer
Kernel SVM trainer using a simplified SMO algorithm.
LassoReg
Lasso (ℓ₁) regularization: min_h L(h) + λ‖h‖₁.
MutualInformation
Mutual information I(X;Y) = H(X) + H(Y) - H(X,Y).
OnlineGradientDescent
Online Gradient Descent with regret bound O(√T).
PACBayesGeneralization
PAC-Bayes generalization bound computation.
PACLearner
PAC learner: wraps accuracy/confidence parameters.
Perceptron
Online Perceptron classifier.
RademacherComplexity
Rademacher complexity estimate for a finite hypothesis class.
RegretBound
Regret bound summary.
SVMClassifier
Support Vector Machine classifier.
SampleComplexity
Sample complexity for PAC learning (Blumer et al. / Vapnik-Chervonenkis bound).
TikhonovReg
Tikhonov (ridge) regularization: min_h L(h) + λ‖h‖².
UCBBandit
UCB1 (Upper Confidence Bound) algorithm for multi-armed bandits.
UniformConvergence
Uniform convergence checker.
VCDimension
VC dimension and growth function calculations.

Enums§

GBLoss
KernelType
Available kernel types.
SVMKernel