I have been dosing off since I began reading and also watching tutorials, but with his, am home🤗🤗🤗🤗 Thanks sir, and I love your speech tone, for it favours all
Hi Thanks for the amazing video At 25:09 when you say "when its a failure the algorithm should backtrack", which algorithm did you mean? The forward checking algorithm or the overall backtracking-CSP algorithm? I am thinking the following We need to try all possible combinations of values in forward checking, and for those combinations where we get consistent assignments we have those values as options in the domain(since we have found out atleast 1 combination that respects it) But suppose in the unfortunate event, we dont event get a single combination that is consistent, then we say to the main algorithm that the domain cannot be updated simply because "we have looked into the future and there is possibly no way what you have chosen is correct so please unassign your current value and try again. Sorry!". Am i remotely right?? Thanks
I have a doubt at 16:57 , It will first remove the assignment and inference and then iterate over the next value right? But you mentioned it as other way around?
In the final table when you told about the number of iterations taken by different techniques for different problems,I dont understand how backtracking couldn't find a solution after all backtracking will ensure visiting all possible states in the tree and so if Min Conflicts could come up with a solution, backtracking will for sure also end up at atleast solution state in the tree,right?
theoretically yes it should find a solution since back tracking always finds a solution. My only guess is that the Tree is so huge that it makes no sense to keep the algorithm running hoping that it'll find a solution
What would be your advise to a student who does not come from CSE background and would like to study AI as a university course. How should he prepare himself and what are the basics should he know beforehand.
1) Learn some basic programming in any language. 2) Study data structures, e.g stack, queue, heap, linked lists and trees (very important) and others (perhaps do an online course on data structures). Implement all of these in a programming language. 3) Learn basic search algorithms like tree traversals e.g breath first and depth first. Understand these completely and program them. Also study and program sorting algorithms. (Perhaps do an online course on these). 4) Brush up on your math, you will need to understand basic set theory, some probability and logical statements. Then you should be good to go.
I have been doing all the above points you mentioned. Interestingly my professor gave me almost the exact directions you have given. I am from EEE but tor the two elective courses that we have to register, i went for Artificial Intelligence and Machine Learning, at the same time! I am at the middle of the semester now. Lets see how I do at the end. I highly appreciate you for taking your time and giving me proper advise.
My being sloppy. What I meant is that for every little square there are 80 potential constraints if I don't yet understand the game very well (that is, each square compared or constrained against all the others). Now, in SODUKU specifically, each square would have 20 relevant constraints without using AllDiff. 8 constraints along the rows, 8 along the columns and 4 left within the bigger squares. Sorry for the sloppiness.
I must say that you are an incredibly good instructor, very clear with great examples. I like how you use several examples. Thank you
7:26 quick maths
minus one
@@MsaybOMaylthats 3
Awesome lectures, you should definitely consider to construct a course on the Russel Norvig book, it would be great :)
Awsome! You saved me from my exam-failure in two days :D
*in 2 hours haha
Lawrence Fulton 2 minutes*
@@navjotsingh2251 2 hours ago :(
I have been dosing off since I began reading and also watching tutorials, but with his, am home🤗🤗🤗🤗
Thanks sir, and I love your speech tone, for it favours all
MashaAllah u cover all the things in a single video and it's awesome! keep uploading .
Amazing content, subscribed. Way clearer than our professor at my university.
thanks , I was stuck with the carry in cryptarithmetic puzzels and you explain it well .
best video for CSP on youtube, thanks a lot!
6:29 'O' Wins the Tic-Tac-Toe game
Hmara madhabanda tumhi ko taapta hai.. tu mast kaam krta h
Two plus two is four, minus one that's three, quick maths! 7:40
Great work! Really introduced the material in an intuitive way with relevant examples
Thank you so much! The examples are really helpful in explaining.
Fantastic teacher. thank you so much
Thanks a billion sir,
tomorrow is my AI paper. you save me :))
Really great Video. Thanks for sharing!!
Thanks Francisco this was very helpful
thanks a lot for easy explanation, it helped
Very clear and understandable.
Hi Thanks for the amazing video
At 25:09 when you say "when its a failure the algorithm should backtrack", which algorithm did you mean? The forward checking algorithm or the overall backtracking-CSP algorithm?
I am thinking the following
We need to try all possible combinations of values in forward checking, and for those combinations where we get consistent assignments we have those values as options in the domain(since we have found out atleast 1 combination that respects it)
But suppose in the unfortunate event, we dont event get a single combination that is consistent, then we say to the main algorithm that the domain cannot be updated simply because "we have looked into the future and there is possibly no way what you have chosen is correct so please unassign your current value and try again. Sorry!".
Am i remotely right??
Thanks
thenkyou , great teaching
Thanks...it's helped me
very good video!
i really need this powerpoint presentation and fast
Great video! Thanks a lot!
Thank you for these awesome videos :)
Thanks a lot. It was really helpful for me:)
I have a doubt at 16:57 , It will first remove the assignment and inference and then iterate over the next value right? But you mentioned it as other way around?
Thank you! Good job
Good video ,Thank you.
Thank you you was so helpful thank you thank you thank you 👍👍👍👍🌹
In the final table when you told about the number of iterations taken by different techniques for different problems,I dont understand how backtracking couldn't find a solution after all backtracking will ensure visiting all possible states in the tree and so if Min Conflicts could come up with a solution, backtracking will for sure also end up at atleast solution state in the tree,right?
theoretically yes it should find a solution since back tracking always finds a solution. My only guess is that the Tree is so huge that it makes no sense to keep the algorithm running hoping that it'll find a solution
Thanks, that was really helpfull.
*Only 18* 👇👇👇
315231.loveisreal.ru
What would be your advise to a student who does not come from CSE background and would like to study AI as a university course. How should he prepare himself and what are the basics should he know beforehand.
1) Learn some basic programming in any language.
2) Study data structures, e.g stack, queue, heap, linked lists and trees (very important) and others (perhaps do an online course on data structures). Implement all of these in a programming language.
3) Learn basic search algorithms like tree traversals e.g breath first and depth first. Understand these completely and program them. Also study and program sorting algorithms. (Perhaps do an online course on these).
4) Brush up on your math, you will need to understand basic set theory, some probability and logical statements.
Then you should be good to go.
I have been doing all the above points you mentioned. Interestingly my professor gave me almost the exact directions you have given. I am from EEE but tor the two elective courses that we have to register, i went for Artificial Intelligence and Machine Learning, at the same time! I am at the middle of the semester now. Lets see how I do at the end. I highly appreciate you for taking your time and giving me proper advise.
How is the no of possibilities for each square is 80? Video at 7:06.
My being sloppy. What I meant is that for every little square there are 80 potential constraints if I don't yet understand the game very well (that is, each square compared or constrained against all the others). Now, in SODUKU specifically, each square would have 20 relevant constraints without using AllDiff. 8 constraints along the rows, 8 along the columns and 4 left within the bigger squares. Sorry for the sloppiness.
really awesome !!
In which playlist you have included this video? I can't find that playlist.
can you please create a video about the chapter 4 of artificial intelligence a modern approach 2nd edition. btw youre a great teacher :)
Check my videos on "Hill Climbing 8 queens" and "simulated annealing" They talk about that.
hi sir! can i have your presentation in this topic :) i just need it for my report
go to fid.cl/ and select artificial intelligence from the courses.
awesome video!
Wait you are allowed to make CS videos in a understandable accent???
waaohh!! that is fantastic
can i get the slides?
Thanks
thank you
Poor sound Qquality
Im ukrainian and drink tea right now :)
How is the situation in Ukraine
@@seunjonathan Im Ukrainian and drink tea in the shelter
You jump a lot of things.
Hard to understand for people that have come to really make sense of the csp's
I don't like the video