PropagatorChannelWeights

class madspace.PropagatorChannelWeights(self: madspace._madspace_py.PropagatorChannelWeights, topologies: collections.abc.Sequence[madspace._madspace_py.Topology], permutations: collections.abc.Sequence[collections.abc.Sequence[collections.abc.Sequence[SupportsInt]]], channel_indices: collections.abc.Sequence[collections.abc.Sequence[SupportsInt]])

Bases: FunctionGenerator

Local multi-channel weights from propagator denominators.

Implements the single-diagram-enhanced (SDE) multi-channel weights \(\alpha_i(x)\) of Eq. (2.7) of [1]: each channel weight is proportional to the product of the Breit-Wigner propagator denominators of its topology, evaluated at the current phase-space point [2, 3]. Only the forward mappings are needed, not their inverses (Sec. 2.1.1 of [1]).

batch is the leading batch dimension. c indexes the channels.

Arguments

• momenta – float, shape (batch, n, 4) – all momenta of the event.

Returns

• channel_weights – float, shape (batch, c) – the normalized per-channel weights \(\alpha_i(x)\).

References

• [1] T. Heimel, O. Mattelaer, R. Winterhalder, “MadSpace”, https://arxiv.org/abs/2602.06895 (Sec. 2.1.1)

• [2] F. Maltoni, T. Stelzer, “MadEvent: automatic event generation with MadGraph”, https://arxiv.org/abs/hep-ph/0208156

• [3] O. Mattelaer, K. Ostrolenk, “Speeding up MadGraph5_aMC@NLO”, https://arxiv.org/abs/2102.00773

Parameters:

• topologies – One Topology per channel.

• permutations – Per channel, the outgoing-particle permutations to average the weight over.

• channel_indices – Per channel, the global channel indices its weight contributes to.