But the spider web itself is implemented using an ad hoc solver of Hooke's law - the solver iteratively moves each free node in the direction of its net force. That is why adding a radius visibly shifts the hub and deforms the frame rather than simply drawing a new straight segment.
![Right, you don't need error bars on error bars.
Probabilistic uncertainty about uncertainty collapses. This is the “monadic join” in probability.
Instead of a coin with random bias p ∼ π, you can flip a coin with the deterministic bias μ.
Just take μ = E[p].](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:weizynbs42yzl4y6de3z4dvu/bafkreigil27ygwcqjtct6wyrjfj7ls7tr3qgpgrbt5n3tfincpg73gas2m)
![For the applicability of Bayes' theorem, it is important when the information $B$ you condition on is actually available.
Let $O \subset B$ be the event of observing $B$ with $P(O)>0$. If $O$ and $A$ are independent with respect to $P_{|B}(\cdot) = P(\cdot \mid B)$,
then
\[
P(A\mid O) = P(A \mid B).
\]](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:weizynbs42yzl4y6de3z4dvu/bafkreihnkdxbvmdk4um6how7qgvurwhfhrkqhlu4baqlnkeepd7z4kkjfe)
![Screenshot:
julia> dag = digraph([U=>X, U=>Z, V=>Y, V=>Z, X=>Y, Z=>X])
DiGraph([1 => 3, 1 => 5, 2 => 4, 2 => 5, 3 => 4, 5 => 3])
julia> observed = [X, Y, Z];
julia> do!(dag, Z)
julia> dag
DiGraph([1 => 3, 2 => 4, 3 => 4, 5 => 3])
julia> adjustments = collect(list_covariate_adjustment(dag, X, Y, ∅, observed))
2-element Vector{Any}:
Set{Int64}()
Set([5])](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:weizynbs42yzl4y6de3z4dvu/bafkreigcizgc6fwacmjofij4yg4xva2ln7sb7aq2ep5zw4lp7ki5qxqr74)
