A* (A Star) Search Algorithm - Computerphile

Super sneaky is from now on totally technical term 😂

who are you talking to

thank you teacher

Okay, what about if you went from Node C to Node E? By my understanding the algorithm would take you via C, L, J, K, E, however, it appears the fastest travel would actually be C, S, B, H, G. It seems like that would be like starting on one end of a city and having to reach the other, and Google Maps is telling you the fastest way to go is straight through the middle of the city, however there's a motorway that travels around the city and is much faster.

Love the content, but it's extremely super difficult to stay focused with camera jumping on and off the paper with the graph. I feel a little bit dizzy every 3 minutes, so had to take often breaks

more I watch this video less I understand A* search

makes no sense to me.. help XD

Thanks!!! Extraordinary clear. Loved the heuristic device!

I would like a video about the D* algorithm, please.

are u a lecturer in nottingham uni?

Excellent explanation. I noticed D, and played it twice to make sure it is a mistake. And, then I read the description. Thank you for putting the video up.

Too fast British accent! I could hardly follow two consecutive sentences.

This example is not admissible (as well as not consistent), right? because the heuristic of 10 overestimates the actual resulting path cost of 9. Is that correct?

You got coffee, Mike?
Really?
Not tea?
Really?

Brilliant

Excessively verbose and in need of an editor. Get your cards in order and ready to use before filming, not during.

i think you mad a mistake regarding D, where f(n) is 14 not 12, thanks

This explanation is great in that it explains what the machine does. Meanwhile I also like the intuition given in polylog's video.

brilliant. thank you for this lecture. So much better than what I had heard thus far.

your face looks like someone punched from the left

support ist kein mord

One neat trick is to "prefer" one metric over another and use power notation to calculate overall heuristic. E.g. a node with distance 7 but weight 2, we added them as 7+2 = 9. But instead of that if we prefer shorter distance over smaller weight then weight should be the base raised to the power of distance.

So this way we can choose easily between two nodes that would otherwise yield the same heuristic if we add them but with the new rule if one node is with weight of 2 and distance of 7 (2^7=128) and another has distance of 2 and weight of 7 (7^2 = 49) ... we chose the later as 49 < 128 because we preferred the one closer to the end node.

Google maps often use this trick.

How to find a better heuristics is fun, you could try different ones. It is very interesting.

well explained dr Pound, sir in my problem I need to determine a set of feasible paths to use in multi-path routing but the main issue is the cost of edges and nodes in the graph it has a quadratic relation to the flow which is unknown in the first place so how can I so how can solve it, please!

Great content and excellent editing!

Ah, the accent from which America stole all of its hard R’s.

I am here after Q* from OpenAI 😂

Of course this pops up on my suggestions after talk of Q* is blowing up. Lol

please try to delete the accent so everyone can understand

lol tape measuring inches

His calculation for "D" in A* was off. D was S + B + D (0 + 2 + 4) or 6, and it had a heuristic of 8, so that is 14. He wrote down 12 in black. Not a major deal breaker here obviously, but just pointing it out b/c that's what us programmers do :). Thanks so much for the video!

If the distance between G and E is set to 15. This could happen, if this route was crossing a mountain, whit lots of bending roads up and down a mountain. It could be close in term of the euclidean distance, but the route could be much longer, both in time, fuel consumption and km/miles. So again if the distance between nodes G and E is set to 15, then this algorithm is not finding the shortest path. Because it will still find this route S -> B -> H -> G -> E even though there is a shorter route, which is S -> C -> L -> J -> K -> E. This happens because G is on the top of the 'ordered list of combined length', just before it finds the solution going to E, which is heavily penalized on the last stretch between G and E. So to mitigate this problem the algorithm should continue, to search after a shorter route, after finding the first solution. All nodes witch is being evaluated, after finding the first 'shortest path', should not be inserted again or deleted into/from the 'combined length sorted list', if the combined length calculated, is longer than the first found 'shortest path'. Then continue until the 'combined length sorted list' is empty. 'shortest path' could be lowered many times.

Genii are left handed.

I guess distance here isn't the scale distance but the number of nodes coming in the middle 😅 or am I wrong?

This new office episode looks great

doesnt understand a word he says!

Student here, thank you! This is a really clean explanation and you clearly really love A star :)

He made a mistake when he expanded D from point B. He should get a heuristic value 14 not 12 (because 2 + 4 + 8, while he just added up 4 + 8 to get the value)

Peregrine ? 🙂

Just rewatched this because I need to solve a maze and didn't understand how a* works.
Time to do some programming

Imagine video games without A* by Peter Hart, Nils Nilsson, and Bertram Raphael - it would just be NPCs running into walls and obstacles.

Why did not I refer this elegent video of your's when I was back in college. I have feared this A* algo till this day, because of the heuristic function, and literally nobody has been able to explain it in this simple manner. Thanks for the video, much appreciated!

Thanks, from Brasil

Always happy when studying for an exam, and Computerphile has made a video about the topic

great and fascinating explanation of a star.

I think you calculate the 'd' node value wrong
it should be 14 instead of 12

2024 Advent of Code brought me here. ❤