Lecture C6: Graphs
Response to 'Muddiest Part of the Lecture Cards'
(8 respondents)
1) Could the graphs be analogous to Markov chains and then would the weights be
analogous to the transitions probabilities?
The Markov chain can be represented as an infinite graph, with probabilities on the
transitions instead of weights.
2) Could the shortest path problem also be approached with Markov chains? Maybe for
systems with more uncertain weights? Or for systems where future weights are unknowns
but estimates exist?
Yes you can.
3) Can you find the shortest path (or least costly) using the adjacent matrix?
Yes, there exists many different kind of algorithms to solve the shortest path problem in
graphs. It is common that the algorithms take an NxN adjacency matrix as input and
calculates an NxN matrix S, where Si,j is the length of the shortest path from vi to vj, or a
special value, e.g., ‘∞’ if there is not path.
4) What is an example of a graph with negative weights?
An interesting question, there are some paths that are undesirable but still have to be
considered (say for instance in the case of a traffic jam), in that case, the weight assigned
to the edge is negative (the goal in this case would be maximize the cost of traversing the
graph).
5) How do you represent ‘
∞
’ in Ada?
The Ada language does not define a representation for infinity. That is left upto the
implementation. You can define your own value for infinity. In the case of Dijkstra’s
shortest path algorithm that we saw in class today, you can use a –ve value to represent
infinity (because the algorithm only works for positive weights and tries to minimize the
total weight).
6) No idea what the problem was with the goose and grain.
An astronaut wants to move herself, a goose, a grain and a fox across a river, from the
north side to the south side. Unfortunately her Rover is so small that she can at most
bring one of the fox, grain or goose across on any trip. Even worse, the fox will eat the
goose if they are left alone on one side of the river, and just as bad, the goose will eat the
grain if they are left alone on one side of the river, so the Astronaut must not leave the
fox alone with the goose nor the goose alone with the grain. How should the Astronaut
solve the problem?
docsity.com
