Compute a status-aware soft trade-off path

Usage

compute_tradeoff_curve(
  J,
  K_target,
  w1_target = NULL,
  lambda_seq = seq(0.1, 1, by = 0.1),
  max_iter = 100L,
  M = .QUAD_NODES_DEFAULT,
  verbose = FALSE,
  loss_type = NULL,
  target = NULL,
  M_verify = NULL,
  log_bounds = .LOG_BOUNDS_DEFAULT,
  control = list(),
  allow_approximate = TRUE
)

Arguments

J: Design size.
K_target: List with mu_K and var_K.
w1_target: Optional legacy W_SB target, translated to an explicit equality target.
lambda_seq: Unique values satisfying 0 < lambda <= 1.
max_iter, M, M_verify, log_bounds, control: Soft solver controls.
verbose: Show path progress.
loss_type: Deprecated curve argument. Only NULL is accepted; the v2 curve always uses fixed input-derived scales.
target: Preferred explicit named weight target.
allow_approximate: Retain approximate points without making them usable warm starts.

Value

A deterministic dpprior_tradeoff_curve data frame with retained fit and condition lists as attributes. No lambda is selected as best.