delvingbitcoin
Estimating Likelihood for Lightning Payments to be (in)feasible
Posted on: June 26, 2024 16:59 UTC
The conversation delves into the intricate aspects of comparing different probability models within the context of minimum cost flow (mcf) computations and wealth distribution feasibility.
The correspondence begins with an acknowledgment of a miscount by the sender, correcting the number of states to ten, thus invalidating their previous counterexample. This correction paves the way for a deeper analysis of the underlying assumptions in their comparison of probability models.
The sender expresses gratitude for the recognition of their point regarding the comparison of different probability models. The discussion centers on whether it is merely more challenging or outright impossible to identify a scenario where the minimum cost flow computation yields a higher probability than the feasibility of certain wealth distributions. Importantly, the sender clarifies that their approach to minimum cost flow computation does not initiate from network states. Instead, it relies on an assumption made in a referenced paper, suggesting that liquidity is uniformly and independently distributed across each channel, which can be found in their publication here.
This premise raises a thought-provoking question about the equivalence of equally weighting all network states to the act of uniformly and independently choosing channel balances. The implication is that there's a foundational assumption in their methodological approach that significantly influences the comparison and interpretation of probability models in the study of network dynamics and economic distributions.