|
aha oh god, i'd have nightmares about hand A* too
|
# ? Sep 20, 2012 15:53 |
|
|
# ? May 9, 2024 16:36 |
|
Graph theory is awesome, got a bit of it and some basic algo's in an intro course that helped decide my major, now using it in another course where our first problem set was tons of applications for social networks. shits tite whiteboard chat: My apartment has a bunch of like, 5 x 5 (feet) whiteboards next to the couches and beds and stuff. So convenient i feel like zuckerberg!!!
|
# ? Sep 20, 2012 15:58 |
|
other than learning how they work, is there really any reason to do graph algorithms (and most other ones) by hand ever?
|
# ? Sep 20, 2012 16:01 |
|
Ronald Raiden posted:other than learning how they work, is there really any reason to do graph algorithms (and most other ones) by hand ever? uh maybe if you're into analog debugging or sth the doing by hand part of the deal helps me find where i gently caress up
|
# ? Sep 20, 2012 16:03 |
|
ZYNGA STOCK CRASHER posted:So convenient i feel like zuckerberg!!! it is NEVER good to feel like zuckerberg
|
# ? Sep 20, 2012 16:04 |
|
Paracetamol Boy posted:uh maybe if you're into analog debugging or sth I guess, its just that, doing anything but small trivial cases by hand is prohibitively difficult/time consuming for a lot of things.
|
# ? Sep 20, 2012 16:05 |
|
abraham linksys posted:dumbest question, but what kind of projects could this be used for? this http://www.faculty.idc.ac.il/arik/SCWeb/vidret/index.html
|
# ? Sep 20, 2012 16:05 |
|
Ronald Raiden posted:I guess, its just that, doing anything but small trivial cases by hand is prohibitively difficult/time consuming for a lot of things. yeah if you want to do hard cases by hand for 80% of applications prob better to cut out the middleman and apply nipple clamps but you can reduce the problem extrapolate some stuff from the simple cases (i don't do it a lot)
|
# ? Sep 20, 2012 16:08 |
|
Ronald Raiden posted:other than learning how they work, is there really any reason to do graph algorithms (and most other ones) by hand ever? prof's excuse was basically that you should do it by hand to gain a 'deeper appreciation' of the algorithm
|
# ? Sep 20, 2012 16:09 |
|
syntaxrigger posted:prof's excuse was basically that you should do it by hand to gain a 'deeper appreciation' of the algorithm other than that excuse
|
# ? Sep 20, 2012 16:10 |
|
Ronald Raiden posted:other than learning how they work, is there really any reason to do graph algorithms (and most other ones) by hand ever? its 1% learning how they work because it might be useful in the future and 99% "im a math prof so i think you should have to know this"
|
# ? Sep 20, 2012 16:10 |
|
did the acm algorithms competitions in college my team was basically my good friend and i taking turns navigating/driving, and the weirdo chinese kid that loved to hand-crank algorithms but couldn't sling java
|
# ? Sep 20, 2012 16:11 |
|
Shaggar posted:its 1% learning how they work because it might be useful in the future and 99% "im a math prof so i think you should have to know this"
|
# ? Sep 20, 2012 16:12 |
|
Cocoa Crispies posted:did the acm algorithms competitions in college for some programming comp we had a maths and graph theory guy, a combinatorics and dp guy, general purpose coding guy and uh team strategist
|
# ? Sep 20, 2012 16:14 |
|
ZYNGA STOCK CRASHER posted:now using it in another course where our first problem set was tons of applications for social networks. shits tite vaguely related http://en.wikipedia.org/wiki/Friendship_paradox neat but pretty obvious if you think about it the applications wrt disease are cool
|
# ? Sep 20, 2012 16:16 |
|
Paracetamol Boy posted:for some programming comp we had a maths and graph theory guy, a combinatorics and dp guy, general purpose coding guy and uh team strategist the ideas guy
|
# ? Sep 20, 2012 16:17 |
|
also someone posted this somewghere idk it might have been this thread and it seemed neat: http://networkx.lanl.gov/
|
# ? Sep 20, 2012 16:18 |
|
Milkie Galore posted:vaguely related wikipedia posted:It can be explained as a form of sampling bias in which people with greater numbers of friends have an increased likelihood of being observed among one's own friends. In contradiction to this, most people believe that they have more friends than their friends have.[2] not me i have no friends and i know it
|
# ? Sep 20, 2012 16:19 |
|
same
|
# ? Sep 20, 2012 16:20 |
|
all my friends believe they have no friends
|
# ? Sep 20, 2012 16:21 |
|
Paracetamol Boy posted:for some programming comp we had a maths and graph theory guy, a combinatorics and dp guy, general purpose coding guy and uh team strategist we were never very good (ucf up the road had some lifers that practiced for hours each day, we did a problem a week), my goal was just to beat the other usf teams (we did)
|
# ? Sep 20, 2012 16:27 |
|
Cocoa Crispies posted:we were never very good (ucf up the road had some lifers that practiced for hours each day, we did a problem a week), my goal was just to beat the other usf teams (we did) i am an avg coder but i am pretty much the cs department spare for when programming competitions have idiotic affirmative-action requirements so i get to work with people who are multiple icpc finalists and it's pretty neat when we get to blaze the competition. it is a bad idea to cramp the style of your imba teammates so usually i just do problem strategies and sort the problem sets to the relevant people and a rundown so they don't have to read or concentrate on team dynamics leaving them free to sperg to their fullest potential i do gcj sometimes but just do the problems on my own i have a pretty huge inferiority complex from working with geniuses
|
# ? Sep 20, 2012 16:37 |
|
if they're so smart why are they in academia
|
# ? Sep 20, 2012 17:04 |
|
ohhh snap
|
# ? Sep 20, 2012 17:06 |
|
i think A* is just a breadth-first search but i cant really remember
|
# ? Sep 20, 2012 17:10 |
|
A* is dijkstra's but with a heuristic for what nodes are closer
|
# ? Sep 20, 2012 17:12 |
|
my prof made us do a USA-scale TSP just to drive home the point of "see its kinda hard to brute force right????" and then a "ok now boot up STATEOFTHEARTSOFTWARE and do it in 2 seconds"
|
# ? Sep 20, 2012 17:14 |
|
ZYNGA STOCK CRASHER posted:my prof made us do a USA-scale TSP just to drive home the point of "see its kinda hard to brute force right????" and then a "ok now boot up STATEOFTHEARTSOFTWARE and do it in 2 seconds" your prof is the worst person
|
# ? Sep 20, 2012 17:16 |
|
yaoi prophet posted:A* is dijkstra's but with a heuristic for what nodes are closer so which one is breadth-first
|
# ? Sep 20, 2012 17:20 |
|
Paracetamol Boy posted:your prof is the worst person but then he moved into the kind of heuristics that those programs use and it was actually kinda enjoyable so that prof is the man thank you, best 1 ive had also im a nerd in a school full of nerds so i loved that poo poo. plus i actually remember poo poo from that class so A+ all around prof
|
# ? Sep 20, 2012 17:21 |
|
rotor posted:so which one is breadth-first
|
# ? Sep 20, 2012 17:22 |
|
rotor posted:so which one is breadth-first breadth first is when you completely ignore the lengths of the edges, dijkstra's reduces to BFS when all the edges have the same weight
|
# ? Sep 20, 2012 17:22 |
|
ZYNGA STOCK CRASHER posted:but then he moved into the kind of heuristics that those programs use and it was actually kinda enjoyable so that prof is the man thank you, best 1 ive had yeah i was kind of imagining him just stopping there that would be a hilariously dick thing to do my department is awesome
|
# ? Sep 20, 2012 17:24 |
|
rotor posted:so which one is breadth-first dijkstra's is bfs with a priority queue instead of a regular queue
|
# ? Sep 20, 2012 17:33 |
|
what's ur fav A* variant mine is jump point search
|
# ? Sep 20, 2012 17:50 |
|
lifelong planning A*
|
# ? Sep 20, 2012 17:52 |
|
aaag i always think 'dfs' when i hear 'dijkstras algorithm'
|
# ? Sep 20, 2012 17:53 |
|
bread first
|
# ? Sep 20, 2012 17:57 |
|
ffs
|
# ? Sep 20, 2012 17:57 |
|
|
# ? May 9, 2024 16:36 |
|
rotor posted:aaag i always think 'dfs' when i hear 'dijkstras algorithm' more like dfo
|
# ? Sep 20, 2012 18:01 |