You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Instead of the basic Dijkstra approach, it could be better to create routes that provide as much entropy as possible. If the flow information contains a very improbable edge, it might be a better idea to split this into several (at this edge: parallel) flows so that failures provide more helpful information about the network. @renepickhardt may have some ideas of how this could be done.
The text was updated successfully, but these errors were encountered:
As discussed I have unfortunately nothing concrete that I can say. I would try discect the flow in a way that the expected information gain of the sent onions is maximzied. (similarly how the optimization goal could already be changed in flow computation). The new optimization goal seems computationally much harder to me which is why I hope it would at least work on the resulting flow graph as that is usally much much smaller than the lightning network
Instead of the basic Dijkstra approach, it could be better to create routes that provide as much entropy as possible. If the flow information contains a very improbable edge, it might be a better idea to split this into several (at this edge: parallel) flows so that failures provide more helpful information about the network. @renepickhardt may have some ideas of how this could be done.
The text was updated successfully, but these errors were encountered: