Greedy Descent
NonArchimedeanMachineLearning.GreedyDescentConfig — Type
GreedyDescentConfig(; strict=false, degree=1, start_branch=1)Configuration for greedy descent optimization.
Fields
strict::Bool: If true, descend one coordinate at a time; if false, descend all coordinatesdegree::Int: Number of children to explore per polydisc node (non-strict mode)start_branch::Int: Coordinate to descend along first (strict mode)
Example
config = GreedyDescentConfig(strict = false, degree = 1)
optim = greedy_descent_init(param, loss, config)NonArchimedeanMachineLearning.greedy_descent — Method
greedy_descent(loss::Loss, param::ValuationPolydisc{S,T,N}, next_branch::Int, settings::GreedyDescentConfig) where {S,T,N}Perform one step of greedy descent optimization.
Computes children of the current parameter and selects the child that minimizes the loss. Can operate in strict mode (one coordinate at a time) or full mode (all coordinates).
Arguments
loss::Loss: The loss function structureparam::ValuationPolydisc{S,T,N}: Current parameter valuesnext_branch::Int: Index of next branch to descend (in strict mode)settings::GreedyDescentConfig: Configuration for greedy descent
Returns
Tuple{ValuationPolydisc{S,T,N}, Int, Bool}: New parameters, next branch index, and convergence status
NonArchimedeanMachineLearning.greedy_descent_init — Method
greedy_descent_init(param::ValuationPolydisc{S,T,N}, loss::Loss, settings::GreedyDescentConfig=GreedyDescentConfig()) where {S,T,N}Initialize an optimization setup for greedy descent.
Arguments
param::ValuationPolydisc{S,T,N}: Initial parameter valuesloss::Loss: The loss function structuresettings::GreedyDescentConfig: Configuration controlling descent behavior
Returns
OptimSetup: Configured optimization setup for greedy descent