Difficult:
The most difficult part is keeping track of the indices and how they affect the equations. Memorization doesn't seem like the most practical approach. As I learn to visualize these better I think I will be able to use them out of intuition rather than memorized steps.
Interesting:
The complexity of the fast version of this algorithm is very impressive. Something that I have found fascinating in this class is learning of all the logical ways to make an algorithm run faster with less computational effort. Most of the time, like in this chapter, it requires a clever way of going about solving the problem. I am motivated to find a faster version of everything I code because of this class.
Saturday, 8 December 2018
Thursday, 6 December 2018
Section 8.7: Haar Wavelets
Difficult:
Conceptually the way we approximate functions with Haar scaling function is easy to understand. Some of the notation and manipulation of these concepts are more complex. Sadly because I am still trying to understand notation that the rest of the chapter depended on, it made it more difficult to understand.
Insightful:
It is cool how well these functions can fit discontinuous functions. This is why it is superior at approximating discontinuous functions than fourier transforms. Which method would be better to approximate nonoscillating continuous functions? are there more types of functions that would not be efficient to use the two methods already studied?
Conceptually the way we approximate functions with Haar scaling function is easy to understand. Some of the notation and manipulation of these concepts are more complex. Sadly because I am still trying to understand notation that the rest of the chapter depended on, it made it more difficult to understand.
Insightful:
It is cool how well these functions can fit discontinuous functions. This is why it is superior at approximating discontinuous functions than fourier transforms. Which method would be better to approximate nonoscillating continuous functions? are there more types of functions that would not be efficient to use the two methods already studied?
Tuesday, 4 December 2018
Section 8.6: Periodic Sampling Theorem
Difficult:
Antialiasing went over my head. But as I understand it we are setting up a new bound that is dependent on n - the number of samples. Then when we compute the equation from the samples we are guaranteed the unique solution. but I think there is more to this.
Insightful:
It is interesting how we filter sound according to human capability. It is interesting how we can have many results from aliasing. I wonder what is the restricting factor for not having enough samples.
Antialiasing went over my head. But as I understand it we are setting up a new bound that is dependent on n - the number of samples. Then when we compute the equation from the samples we are guaranteed the unique solution. but I think there is more to this.
Insightful:
It is interesting how we filter sound according to human capability. It is interesting how we can have many results from aliasing. I wonder what is the restricting factor for not having enough samples.
Saturday, 1 December 2018
Section 8.5: Convolution
Difficult:
I still don't understand the application of convolutions. I am not sure what the graph on figure 8.5.1 is supposed to be representing.
I understand that using Fourier transforms and discrete convolutions we can compute the circular convolution a lot faster. Now it would be nice to know why we are computing it.
Interesting:
I have been wanting to learn more about convolution. I have a vague remembrance of it from math 334 and have seen it come up in research papers. I didn't know they correlated with Fourier transforms. Looking forward to understanding them better.
I still don't understand the application of convolutions. I am not sure what the graph on figure 8.5.1 is supposed to be representing.
I understand that using Fourier transforms and discrete convolutions we can compute the circular convolution a lot faster. Now it would be nice to know why we are computing it.
Interesting:
I have been wanting to learn more about convolution. I have a vague remembrance of it from math 334 and have seen it come up in research papers. I didn't know they correlated with Fourier transforms. Looking forward to understanding them better.
Thursday, 29 November 2018
The Discrete Fourier Transform
Difficult:
I would like to understand why the unbolded w to the power of (k-m)*l is 0 whenever k != m. The calculation is just a summation which I would expect to have a value not equal to 0.
Interesting:
I just computed the matrix W that is shown in algorithm 8.4.1 for the lab. This way of computing the discrete Fourier transforms is a lot quicker than doing a loop.
Filtering noise is a very cool and useful application of Fourier Tranforms. It is interesting that the filtering method that does not focus on the desired frequency range is the one that matches the data the least. I would have expected it to be the other way around.
I would like to understand why the unbolded w to the power of (k-m)*l is 0 whenever k != m. The calculation is just a summation which I would expect to have a value not equal to 0.
Interesting:
I just computed the matrix W that is shown in algorithm 8.4.1 for the lab. This way of computing the discrete Fourier transforms is a lot quicker than doing a loop.
Filtering noise is a very cool and useful application of Fourier Tranforms. It is interesting that the filtering method that does not focus on the desired frequency range is the one that matches the data the least. I would have expected it to be the other way around.
Tuesday, 27 November 2018
Section 8.3: Convergence of Fourier Series
Difficult:
I followed most lemmata and saw how the logic behind the steps. I wish the chapter would have had more application/motivation because I don't really know why we are proving all these things. For example, the last proof that was about a theorem in the first subsection of chapter 8, I don't even know the purpose of that theorem.
Interesting:
i) The graph is cool. It was the highlight, which says more about the this chapter than I have to say with words
ii) the plural of lemma is lemmata
I followed most lemmata and saw how the logic behind the steps. I wish the chapter would have had more application/motivation because I don't really know why we are proving all these things. For example, the last proof that was about a theorem in the first subsection of chapter 8, I don't even know the purpose of that theorem.
Interesting:
i) The graph is cool. It was the highlight, which says more about the this chapter than I have to say with words
ii) the plural of lemma is lemmata
Sunday, 25 November 2018
Section 8.1: Fourier Series. & Appendix B
Difficult:
There was a lot of new information in this reading assignment. It is just hard to know if I will need to solve the problems out of what I read. I think if there was more visual representation of the projections that are happening in the equation it would become more intuitive to use them.
Insightful:
I am looking forward to learning how to use Fourier Series since I've heard it is used extensively in engineering fields. The part on number music theory was also interesting but I am looking to use it on what interests me.
There was a lot of new information in this reading assignment. It is just hard to know if I will need to solve the problems out of what I read. I think if there was more visual representation of the projections that are happening in the equation it would become more intuitive to use them.
Insightful:
I am looking forward to learning how to use Fourier Series since I've heard it is used extensively in engineering fields. The part on number music theory was also interesting but I am looking to use it on what interests me.
Tuesday, 20 November 2018
Section 7.3: Hashing
Difficult:
Conceptually, everything made sense in this section. The only difficult thing I see is deciding which method to use for handling collisions. They all have a potential con but it really does depend on the type of data you are storing.
Insightful:
I like the fact that the probability for having the same birthday came back for something more practical. I also like what we are learning because when I learned that a set had a hash function in it made me want to use sets more in my code. Finding out the usefulness of something can help us get away from comfortable habits.
Conceptually, everything made sense in this section. The only difficult thing I see is deciding which method to use for handling collisions. They all have a potential con but it really does depend on the type of data you are storing.
Insightful:
I like the fact that the probability for having the same birthday came back for something more practical. I also like what we are learning because when I learned that a set had a hash function in it made me want to use sets more in my code. Finding out the usefulness of something can help us get away from comfortable habits.
Sunday, 18 November 2018
Section 7.2: Importance, Inversion, and Rejection Sampling
Difficult:
In this section, we are learning about using different distributions to use the Monte Carlo Methods. We are trying to match it as much as possible to a function that we are using Monte Carlo on. I want to understand the importance of the inverse sampling better because I feel like I did not understand the examples at the end of the subsection.
Interesting:
Rejection Sampling uses two methods. One that can be more accurate but can also diverge. If the first diverges the second, the more reliable method is used. Pretty much an if statement that can help with the approximation of the model.
In this section, we are learning about using different distributions to use the Monte Carlo Methods. We are trying to match it as much as possible to a function that we are using Monte Carlo on. I want to understand the importance of the inverse sampling better because I feel like I did not understand the examples at the end of the subsection.
Interesting:
Rejection Sampling uses two methods. One that can be more accurate but can also diverge. If the first diverges the second, the more reliable method is used. Pretty much an if statement that can help with the approximation of the model.
Thursday, 15 November 2018
Section 7.1: Monte Carlo Methods
Difficult:
I don't think there was anything too difficult about this section but that does not mean I would not have a hard time explaining the concept. The main idea I am getting out of it is that for the outcomes of random variables in a distribution we can compute the expected value of integrals and other models. Like the Riemann sum example where the same amount of points were taken and because of the distribution, we get a pretty good estimation of the integral itself. I am excited to learn about these methods because I just went to a meeting today where this idea came up.
Insightful:
The approximation of pi is remarkable. It is mind boggling how random selected values can give such an accurate response.
I don't think there was anything too difficult about this section but that does not mean I would not have a hard time explaining the concept. The main idea I am getting out of it is that for the outcomes of random variables in a distribution we can compute the expected value of integrals and other models. Like the Riemann sum example where the same amount of points were taken and because of the distribution, we get a pretty good estimation of the integral itself. I am excited to learn about these methods because I just went to a meeting today where this idea came up.
Insightful:
The approximation of pi is remarkable. It is mind boggling how random selected values can give such an accurate response.
Tuesday, 13 November 2018
Test Prep
- Which topics and ideas do you think are the most important out of those we have studied?
- What kinds of questions do you expect to see on the exam?
Calculating probabilities with Bayes' rule
Doing lots of algebra
- What do you need to work on understanding better before the exam?
Saturday, 10 November 2018
Section 6.5: Bayesian Statistics
Difficult:
The unknown paramenters are still a concept I am getting used to. Condition probability with many paramenters and changing outcomes is the idea I am trying to rap my mind around. Fortunately, this section has examples to help with that.
Insightful:
I find it interesting that you can mix two different types of distributions and get a similar or different distribution out. The idea of which one to use is still a little hazy but nothing practice can't teach.
The unknown paramenters are still a concept I am getting used to. Condition probability with many paramenters and changing outcomes is the idea I am trying to rap my mind around. Fortunately, this section has examples to help with that.
Insightful:
I find it interesting that you can mix two different types of distributions and get a similar or different distribution out. The idea of which one to use is still a little hazy but nothing practice can't teach.
Thursday, 8 November 2018
Section 6.3: The Central Limit Theorem
Difficult:
How come in the beta distribution graphs the more likely outcome to the left did not happen more often? I am now assuming the scripted N is the normal distribution simulated by the Central Limit Theorem (since it was not defined).
Insightful:
I don't understand notation but I understand the idea. If we have events happening a lot, they start to simulate a normal distribution. Even if they were drawn from a different distribution.
How come in the beta distribution graphs the more likely outcome to the left did not happen more often? I am now assuming the scripted N is the normal distribution simulated by the Central Limit Theorem (since it was not defined).
Insightful:
I don't understand notation but I understand the idea. If we have events happening a lot, they start to simulate a normal distribution. Even if they were drawn from a different distribution.
Tuesday, 6 November 2018
Section 6:2 The Law of Large Numbers
Difficult:
The theorems in this chapter are fairly straightforward but I am afraid that I have forgotten parts of the last chapter that are used as facts in this chapter. It is great to review but at the same time I feel like there are concepts going over my head even though I am supposed to be understanding them from previous material (for example the variable 1E) . From now on, I will review the previous chapters as I do the homework.
Insightful:
I am starting to see from the material in this call and the volume 1 class that inequalities will be an important part of problem-solving. I can see the bounds on how probable an event is as useful information when making a decision (whether it be business or engineering).
The theorems in this chapter are fairly straightforward but I am afraid that I have forgotten parts of the last chapter that are used as facts in this chapter. It is great to review but at the same time I feel like there are concepts going over my head even though I am supposed to be understanding them from previous material (for example the variable 1E) . From now on, I will review the previous chapters as I do the homework.
Insightful:
I am starting to see from the material in this call and the volume 1 class that inequalities will be an important part of problem-solving. I can see the bounds on how probable an event is as useful information when making a decision (whether it be business or engineering).
Sunday, 4 November 2018
Section 6.1: Estimation
Difficult:
The book could have done a better job at defining certain terms like parameters, or likeliness. The examples seemed more like proofs than real-life examples. I was trying to make my own examples to try to understand the material better but I don't know if my assumptions are correct.
Insightful:
I like the fact that chapter 6 is a continuation of chapter 5 since I feel like more exposure to these terms will make them stick better. Hopefully, everything will make more sense as we get closer to the test.
The book could have done a better job at defining certain terms like parameters, or likeliness. The examples seemed more like proofs than real-life examples. I was trying to make my own examples to try to understand the material better but I don't know if my assumptions are correct.
Insightful:
I like the fact that chapter 6 is a continuation of chapter 5 since I feel like more exposure to these terms will make them stick better. Hopefully, everything will make more sense as we get closer to the test.
Thursday, 1 November 2018
Section 5.7: Multivariate Radom Variables
Difficult:
Everything. This chapter was extremely dense. There is multiply options in a probability space. But the logic for the proves was not easy to follow
Questionable:
Why is the expected value greater than 1 for the different options of the balls example?
Everything. This chapter was extremely dense. There is multiply options in a probability space. But the logic for the proves was not easy to follow
Questionable:
Why is the expected value greater than 1 for the different options of the balls example?
Tuesday, 30 October 2018
Section 5.6: Continuous Random Variables
Difficult:
The last pages on the different types of continuous distributions was challenging to grasp. I will have to refer back to this chapter to see the relationship between the many variables that were introduced.
Insightful:
It is interesting how the formulas we derived for discrete probability carry over to continuous probability. I like the idea that choosing one value from a countable set of things is essentially 0 since there are so many other choices that could happen.
The last pages on the different types of continuous distributions was challenging to grasp. I will have to refer back to this chapter to see the relationship between the many variables that were introduced.
Insightful:
It is interesting how the formulas we derived for discrete probability carry over to continuous probability. I like the idea that choosing one value from a countable set of things is essentially 0 since there are so many other choices that could happen.
Sunday, 28 October 2018
Section 5.5: Discrete Distributions
Difficult:
The distributions were pretty straight forward. There wasn't anything behind the theory of the Poisson Distribution. I think it is really cool but how did he derive it?
Insightful:
I think it is cool that with a probability high enough you can consider certain values essentially not possible. Or on the other hand, if it is low enough, certain outcomes impossible to achieve
The distributions were pretty straight forward. There wasn't anything behind the theory of the Poisson Distribution. I think it is really cool but how did he derive it?
Insightful:
I think it is cool that with a probability high enough you can consider certain values essentially not possible. Or on the other hand, if it is low enough, certain outcomes impossible to achieve
Thursday, 25 October 2018
Section 5.4: Discrete Random Variables
Difficult:
This was not an easy read. I am not used to this notation for a random variable. What I understood is that it is a function that counts each time a desired outcome happens. The law of the Unconscious Statistician went right over my head. The set up for it was really difficult to follow.
Insightful.
I am not sure if I understood the problem for flipping a coin but I thought it was really cool that we a convergent sum we can assume that the coin will flip tails twice before it flips heads.
This was not an easy read. I am not used to this notation for a random variable. What I understood is that it is a function that counts each time a desired outcome happens. The law of the Unconscious Statistician went right over my head. The set up for it was really difficult to follow.
Insightful.
I am not sure if I understood the problem for flipping a coin but I thought it was really cool that we a convergent sum we can assume that the coin will flip tails twice before it flips heads.
Wednesday, 24 October 2018
Section 5.3: Independence, Paradoxes, and Pitfalls
Difficult:
I am not paranoid to fall for the prosecutor's fallacy. I had never worried about this in the stat class I took, mostly because we never talked about it. Everything was pretty straightforward. I know I will need some practice knowing whether a group of events are independent.
Insightful:
I like the plane example where the planes that survived did not need more armor where they had bullet holes since they survived with them. The armor should be placed where there are not that many bullet holes since that is probably the reason why others did not survive. It is also interesting that most of Harvard students got these types of questions wrong.
I am not paranoid to fall for the prosecutor's fallacy. I had never worried about this in the stat class I took, mostly because we never talked about it. Everything was pretty straightforward. I know I will need some practice knowing whether a group of events are independent.
Insightful:
I like the plane example where the planes that survived did not need more armor where they had bullet holes since they survived with them. The armor should be placed where there are not that many bullet holes since that is probably the reason why others did not survive. It is also interesting that most of Harvard students got these types of questions wrong.
Sunday, 21 October 2018
Section 5.2: Conditional Probability and Bayes' Rule
Difficult:
For the chain rule example, I don't see why it would be an intersection between two events and not the union of them. I am having a hard time with the notations of events. Maybe that's where my learning gap is.
Interesting:
We talked in class about how there is no intuitive way to build strategies for solving these problems. I am wondering if this is worth putting extra effort to get good at. I took a stats class and learned most of these methods but I forgot them since. Is this actually practical in the work place or is it a type of skill that will impress your friends? (I am very impressed when people can do these problems)
For the chain rule example, I don't see why it would be an intersection between two events and not the union of them. I am having a hard time with the notations of events. Maybe that's where my learning gap is.
Interesting:
We talked in class about how there is no intuitive way to build strategies for solving these problems. I am wondering if this is worth putting extra effort to get good at. I took a stats class and learned most of these methods but I forgot them since. Is this actually practical in the work place or is it a type of skill that will impress your friends? (I am very impressed when people can do these problems)
Thursday, 18 October 2018
Section 5.1: Probability Theory
Difficult:
I am sure I have seen these types of problems before because I took Stats 201 but the new notation is making it difficult to recall. Other than that it is pretty straight forward.
Insightful:
It is cool that Sterling's Approximation was used to figure out the probability of two people having the same birthday in a room with 'k' people. It says that 365! makes Sterling's Approximation accurate. In Chapter 2 we learned 22026 was a good amount to have a good approximation. So because 22026 < 365! we know it will be accurate?
I am sure I have seen these types of problems before because I took Stats 201 but the new notation is making it difficult to recall. Other than that it is pretty straight forward.
Insightful:
It is cool that Sterling's Approximation was used to figure out the probability of two people having the same birthday in a room with 'k' people. It says that 365! makes Sterling's Approximation accurate. In Chapter 2 we learned 22026 was a good amount to have a good approximation. So because 22026 < 365! we know it will be accurate?
Sunday, 14 October 2018
Section 4.5: Hard Problems
Difficult:
I am having a hard time keeping NP, NP-complete, and Np-hard straight. I am going to need more help than the diagram they have in the book.
Interesting:
It's interesting how what seems like simple problems (simple because as we are doing them we don't over think it and just take action) can have such complicated optimal solutions. Since the complexity increases by a lot every time you add one more option, I would just divide up the work (for example more than one salesman) to the point that it is feasible.
I am having a hard time keeping NP, NP-complete, and Np-hard straight. I am going to need more help than the diagram they have in the book.
Interesting:
It's interesting how what seems like simple problems (simple because as we are doing them we don't over think it and just take action) can have such complicated optimal solutions. Since the complexity increases by a lot every time you add one more option, I would just divide up the work (for example more than one salesman) to the point that it is feasible.
Thursday, 11 October 2018
Section 4.3: Minimum Spanning Trees
Difficult:
I am not versed in algorithm talk yet. I read one paragraph out-loud and considered myself a crazy person. A visual representation would be nice but I think the best way to understand it is by coding the algorithm.
Insightful:
I liked the application of networks and finding the path with the least cost. I am going to start thinking of ways to apply this to find the path of least resistance in life planning (most efficient)
I am not versed in algorithm talk yet. I read one paragraph out-loud and considered myself a crazy person. A visual representation would be nice but I think the best way to understand it is by coding the algorithm.
Insightful:
I liked the application of networks and finding the path with the least cost. I am going to start thinking of ways to apply this to find the path of least resistance in life planning (most efficient)
Wednesday, 10 October 2018
Section 4.2: Graph Search Algorithms
Insightful:
That comic was the insightful thought. What if humans thought as we make computer think? Can we make life more efficient? Would be want to? I am glad we have reason.
Difficult:
My brain is fried from today's homework. But out of the many things I did not understand from this chapter, I would pose the question why would BST give you the shortest path first if it could have been the longest?
That comic was the insightful thought. What if humans thought as we make computer think? Can we make life more efficient? Would be want to? I am glad we have reason.
Difficult:
My brain is fried from today's homework. But out of the many things I did not understand from this chapter, I would pose the question why would BST give you the shortest path first if it could have been the longest?
Sunday, 7 October 2018
Section 4.1: Dynamic Programming
Difficult:
I don't think I understood either section very well. Memoization conceptually made sense but I don't know how to program it. I think Bottom-up makes sense, but I get lost in the details or notation. The coin example would have been simple enough but I just couldn't see how the arguments inside the min function made sense.
Insightful:
I am excited to learn these methods and to find application for them in my concentration. It seems very practical. I think it is cool that we are getting into optimization.
I don't think I understood either section very well. Memoization conceptually made sense but I don't know how to program it. I think Bottom-up makes sense, but I get lost in the details or notation. The coin example would have been simple enough but I just couldn't see how the arguments inside the min function made sense.
Insightful:
I am excited to learn these methods and to find application for them in my concentration. It seems very practical. I think it is cool that we are getting into optimization.
Thursday, 4 October 2018
3.4 Priority Queues and Heaps
Difficult:
I am having a hard time differentiating from all the different complexities for all the different methods. Probably because there is a lot of text between the explanations. It would be useful to see a table with: method, complexity for insterting, complexity for removing, etc.
Interesting:
If I understood correctly, a binary heap is an array sorted as if it was a tree. The kth position has children in the 2k+1 and 2k+2 positions. The priority heap is the tree form of this. There are the terms that are used interchangeably but should not be. This is the reason why I can see why they should not be used interchangeably but the book implies that there are many more.
I am having a hard time differentiating from all the different complexities for all the different methods. Probably because there is a lot of text between the explanations. It would be useful to see a table with: method, complexity for insterting, complexity for removing, etc.
Interesting:
If I understood correctly, a binary heap is an array sorted as if it was a tree. The kth position has children in the 2k+1 and 2k+2 positions. The priority heap is the tree form of this. There are the terms that are used interchangeably but should not be. This is the reason why I can see why they should not be used interchangeably but the book implies that there are many more.
Wednesday, 3 October 2018
Section 3.3: Search Trees
Difficult:
I feel like the book is explaning less and less about Complexity. It could be because they are assuming we are getting the hang of it. But in reality I feel like I am losing the ability to calculate it. For example, are we still working with FLOPs?
AVL Trees went way over my head. I need to understand the unbalance numbers better (like -2 and 2)
Interesting:
I think binary trees are really cool. At first the way the "keys" are indexed might seem arbitrary but there is some logic at least for the children nodes.
I feel like the book is explaning less and less about Complexity. It could be because they are assuming we are getting the hang of it. But in reality I feel like I am losing the ability to calculate it. For example, are we still working with FLOPs?
AVL Trees went way over my head. I need to understand the unbalance numbers better (like -2 and 2)
Interesting:
I think binary trees are really cool. At first the way the "keys" are indexed might seem arbitrary but there is some logic at least for the children nodes.
Sunday, 30 September 2018
Section 3.2: Tree and Tree-Based Data Structures
Difficult:
Some of the statements seem trivial. Which makes it hard to know how to prove them. Maybe it is just that they all sound the same since it is new vocabulary.
I want to understand the calculation of the complexities better. Especially with the linked lists.
Insights:
I like that the labs are complementing the matrial in this book. I liked the visualization data as trees. Trees are undirected and rooted trees are drawn upside down so that there is no need to draw arrows. The arrows are more for the direction the tree grows and not the direction of the edges. This is my hypothesis.
Some of the statements seem trivial. Which makes it hard to know how to prove them. Maybe it is just that they all sound the same since it is new vocabulary.
I want to understand the calculation of the complexities better. Especially with the linked lists.
Insights:
I like that the labs are complementing the matrial in this book. I liked the visualization data as trees. Trees are undirected and rooted trees are drawn upside down so that there is no need to draw arrows. The arrows are more for the direction the tree grows and not the direction of the edges. This is my hypothesis.
Friday, 28 September 2018
Section 3.1: Theory of Graphs
Difficult:
There wasn't anything too difficult in this section. I will have to take a look at Adjacency Matrices tomorrow to understand them better. I think it is really cool how we can represent walks beween vertices.
Insightful:
I really like the graphs. They are like mathematical puzzles. I hope the application can be used through out my work becauset they are kind of fun.
There wasn't anything too difficult in this section. I will have to take a look at Adjacency Matrices tomorrow to understand them better. I think it is really cool how we can represent walks beween vertices.
Insightful:
I really like the graphs. They are like mathematical puzzles. I hope the application can be used through out my work becauset they are kind of fun.
Tuesday, 25 September 2018
Section 2.1: The Gamma Function and Stirling's Approximation
Difficult:
There are a lot of claims with inequalities that I don't find as obvious as the author. This was probably intentional since it is one of the exercises to prove them. This is the most intimidating section so far.
Understanding:
We are giving bounds to functions we are unsure of behavior and trying to reduce this bound as much as possible. The gamma function has very interesting properties stated in Remark 2.1.6.
I spend 5-6 hours on each homework assignment
The lectures and readings do provide the tools to solve the problems. I am very impressed by that since 341 did not feel that way.
It helps to try the problems for 30 minutes by myself then work in a group.
A goal for myself is to start the homework before the lecture.
There are a lot of claims with inequalities that I don't find as obvious as the author. This was probably intentional since it is one of the exercises to prove them. This is the most intimidating section so far.
Understanding:
We are giving bounds to functions we are unsure of behavior and trying to reduce this bound as much as possible. The gamma function has very interesting properties stated in Remark 2.1.6.
I spend 5-6 hours on each homework assignment
The lectures and readings do provide the tools to solve the problems. I am very impressed by that since 341 did not feel that way.
It helps to try the problems for 30 minutes by myself then work in a group.
A goal for myself is to start the homework before the lecture.
Saturday, 22 September 2018
Section 1.10: Divide and Conquer
Difficult:
I haven't coded much with recursion so it all felt very foreign to me. I will need to code a couple of these to know what the function is doing, especially at the return. I will go back to the algorithms in the book after I code a couple of simple recursive functions.
Insightful:
The Master theorem is a method to compute the asymptotical complexity of big-O. I still don't understand the notation for complexity (T(n)).
It is interesting how more operations makes the code run faster and more efficiently.
I haven't coded much with recursion so it all felt very foreign to me. I will need to code a couple of these to know what the function is doing, especially at the return. I will go back to the algorithms in the book after I code a couple of simple recursive functions.
Insightful:
The Master theorem is a method to compute the asymptotical complexity of big-O. I still don't understand the notation for complexity (T(n)).
It is interesting how more operations makes the code run faster and more efficiently.
Thursday, 20 September 2018
Section 1.9: Primes and Remainders
Importance:
Prime numbers are fascinating. I once wrote up an excel sheet with a list of natural numbers and tried to see some patter between them.
Difficult:
Remark 1.9.8 I don't see why all those options are equal. Got it, there is a comma in between separating them. The algorithm for finding Inverses in Zn skips a lot of steps. I can't see what they do with the 31 in the last line. Potentially splitting it up and grouping part together.
Reflexive:
Substitution Rule looks like a fun one you can impress your friends with. Learned a cool new algorithm thanks to it. I had no idea mod would have so much overlap with prime numbers.
Prime numbers are fascinating. I once wrote up an excel sheet with a list of natural numbers and tried to see some patter between them.
Difficult:
Remark 1.9.8 I don't see why all those options are equal. Got it, there is a comma in between separating them. The algorithm for finding Inverses in Zn skips a lot of steps. I can't see what they do with the 31 in the last line. Potentially splitting it up and grouping part together.
Reflexive:
Substitution Rule looks like a fun one you can impress your friends with. Learned a cool new algorithm thanks to it. I had no idea mod would have so much overlap with prime numbers.
Tuesday, 18 September 2018
1.8 Division and Divisors
Importance:
I can see these methods for dividing being useful in solving problems... I still don't see which ones, but I can feel it.
Difficult:
The proof for Division Theorem will require me to go over it again in order to understand it better. I believe is just getting used to a,b,q, and r. Overall these proofs are hard to read. The lemma 1.8.14 was the hardest of them all. I don't understand why we need this new notation.
Reflexive:
Most of this chapter was a review of 290 and concepts that I found fascinating. Like the fact that we still use the Euclidean algorithm because we still haven't found anything faster is really impressive. We have improved many methods for computing certain values, but this one is still the fastest and least computationally heavy and it was discovered 300 B.C. Remarkable.
I can see these methods for dividing being useful in solving problems... I still don't see which ones, but I can feel it.
Difficult:
The proof for Division Theorem will require me to go over it again in order to understand it better. I believe is just getting used to a,b,q, and r. Overall these proofs are hard to read. The lemma 1.8.14 was the hardest of them all. I don't understand why we need this new notation.
Reflexive:
Most of this chapter was a review of 290 and concepts that I found fascinating. Like the fact that we still use the Euclidean algorithm because we still haven't found anything faster is really impressive. We have improved many methods for computing certain values, but this one is still the fastest and least computationally heavy and it was discovered 300 B.C. Remarkable.
Saturday, 15 September 2018
Section 1.7: Products and Counting
Importance:
Products can be useful for probability and shorthand to incredibly long addition. The Pascal triangle is a very useful tool that I want to understand better.
Difficult:
When choosing combinations and rearrangements, the most difficult part is knowing what to put in the denominator. As I understand it, we have any repeats on the denominator and using the factorial function on them. This is when they are not dependent on the other. When they are in sequence then you just multiply the different options which have no need of the equation but it can still be used.
Reflexive:
Taking the idea of the multiplication rule, where the product of the Cartesian product has a cardinality of the cardinality of the sets used for the product. We can know how many possible combinations of phone numbers there can be. Each spot can be a number between 0-9 inclusive. Therefore the cardinality of the different combinations inside the set of all telephone numbers is 10,000,000,000. We are not going to need to bump up an extra digit for a while.
Products can be useful for probability and shorthand to incredibly long addition. The Pascal triangle is a very useful tool that I want to understand better.
Difficult:
When choosing combinations and rearrangements, the most difficult part is knowing what to put in the denominator. As I understand it, we have any repeats on the denominator and using the factorial function on them. This is when they are not dependent on the other. When they are in sequence then you just multiply the different options which have no need of the equation but it can still be used.
Reflexive:
Taking the idea of the multiplication rule, where the product of the Cartesian product has a cardinality of the cardinality of the sets used for the product. We can know how many possible combinations of phone numbers there can be. Each spot can be a number between 0-9 inclusive. Therefore the cardinality of the different combinations inside the set of all telephone numbers is 10,000,000,000. We are not going to need to bump up an extra digit for a while.
Thursday, 13 September 2018
1.5 Nested Loops
Importance:
Most of the coding I've done to solve a wide variety of problems have involved nested loops. We can use sums to compute the complexity of these algorithms.
Difficult:
I read FLOPs like 20-30 times but I am still having a hard time remembering the system. It stands for Floating-Point Operations. This is similar to the temporal complexity then. It does all the basic arithmetic on numbers that are not necessarily integers. After writing this, I feel more confident about FLOPs.
Reflexive:
I am starting to see the importance of leading term coefficients. If there is a part of your code that runs at a degree lower than your leading term it is no longer the place you should be focusing on. I learned about LU decomposition in a numerical methods class. We did not go into the theory very much so I am looking forward to using it more and understanding it better. Very cool how changing the order of the nested loop can affect FLOPs. As well as what is inside the loops, which is not as easy to see in a nested for loop.
Most of the coding I've done to solve a wide variety of problems have involved nested loops. We can use sums to compute the complexity of these algorithms.
Difficult:
I read FLOPs like 20-30 times but I am still having a hard time remembering the system. It stands for Floating-Point Operations. This is similar to the temporal complexity then. It does all the basic arithmetic on numbers that are not necessarily integers. After writing this, I feel more confident about FLOPs.
Reflexive:
I am starting to see the importance of leading term coefficients. If there is a part of your code that runs at a degree lower than your leading term it is no longer the place you should be focusing on. I learned about LU decomposition in a numerical methods class. We did not go into the theory very much so I am looking forward to using it more and understanding it better. Very cool how changing the order of the nested loop can affect FLOPs. As well as what is inside the loops, which is not as easy to see in a nested for loop.
Tuesday, 11 September 2018
Section 1.4: REindexing and Changing Order of Summation
Importance:
At times we can facilitate a summation by changing the bounds of summation.
Difficult:
The tables took a good 10-20 minutes to convince myself that they were the same thing. This was for the last two figures. The first one was not bad. It is just difficult to keep track of the order of operations.
Practice will help me to know what can be done and how it should be done.
Reflexive:
This is a useful method when you have closed-form equation for a series but it doesn't match it exactly. In which case it would be a shortcut to having to re-calculate.
It is interesting how in order to change the summation order we have to change the indices. I am starting to think of them as a nested loop but the idea escapes me at times.
example 1.4.13 does a good job of demonstrating how changing the order of the summations can simplify the problem as stated in the chapter summary.
At times we can facilitate a summation by changing the bounds of summation.
Difficult:
The tables took a good 10-20 minutes to convince myself that they were the same thing. This was for the last two figures. The first one was not bad. It is just difficult to keep track of the order of operations.
Practice will help me to know what can be done and how it should be done.
Reflexive:
This is a useful method when you have closed-form equation for a series but it doesn't match it exactly. In which case it would be a shortcut to having to re-calculate.
It is interesting how in order to change the summation order we have to change the indices. I am starting to think of them as a nested loop but the idea escapes me at times.
example 1.4.13 does a good job of demonstrating how changing the order of the summations can simplify the problem as stated in the chapter summary.
Saturday, 8 September 2018
Section 1.3 Summation
Importance:
The summation of complexity is important when algorithms become complex because it becomes more difficult to calculate all at once.
Difficult:
The definition of the Difference Operator. It seems like the set up for the definition is not used at all during the definition. I am still trying to piece it together. The example in the chapter helps with the understanding of the definition but the set up is still confusing.
Right now the notation is what I am getting caught up on. I feel the same way about the fundamental theorem of finite calculus as I did for the definition of difference operator.
Reflexive:
Very cool how the difference operator represents the derivative of a function. It is a very simple math operation but it represents it well.
Looking forward to the pattern for finding the equations that represent higher order differentiation through the fundamental theorem of finite calculus.
The summation of complexity is important when algorithms become complex because it becomes more difficult to calculate all at once.
Difficult:
The definition of the Difference Operator. It seems like the set up for the definition is not used at all during the definition. I am still trying to piece it together. The example in the chapter helps with the understanding of the definition but the set up is still confusing.
Right now the notation is what I am getting caught up on. I feel the same way about the fundamental theorem of finite calculus as I did for the definition of difference operator.
Reflexive:
Very cool how the difference operator represents the derivative of a function. It is a very simple math operation but it represents it well.
Looking forward to the pattern for finding the equations that represent higher order differentiation through the fundamental theorem of finite calculus.
Thursday, 6 September 2018
Sections 1.1-1.2
Starting with the difficult part of 1.2
In the algorithm 1.2.1 I don't see how one of the lists ends up being empty.
This is teaching us about the equivalence relation (~) between leading-order coefficients. As I understand it there is a way to state simply what is the difficulty of the algorithm. For example 3x^2 -3 ~ 3x^2. We don't really worry about the lower power terms.
The chapter does a good job at showing how the same outcome can be achieved by two different algorithms. It also shows that one is superior by decreasing the big O leading coefficient.
From chapter 1.1 examples 1.1.10 and 1.1.11 were the hardest to understand but the most useful for understanding how to prove these types of problems. These examples were what gave me the tools to solve the homework proof problems.
I found the proposition 1.1.3 very useful in helping undertand the concepts of big-O and little-0.
I am now realizing that I have more to understand about temporal and spatial complexity since I had difficulty explaining part b) of 1.6 and I have my doubts as to whether I understood the implementation during 1.2. I will make sure to ask more question on the subject in the days to come.
In the algorithm 1.2.1 I don't see how one of the lists ends up being empty.
This is teaching us about the equivalence relation (~) between leading-order coefficients. As I understand it there is a way to state simply what is the difficulty of the algorithm. For example 3x^2 -3 ~ 3x^2. We don't really worry about the lower power terms.
The chapter does a good job at showing how the same outcome can be achieved by two different algorithms. It also shows that one is superior by decreasing the big O leading coefficient.
From chapter 1.1 examples 1.1.10 and 1.1.11 were the hardest to understand but the most useful for understanding how to prove these types of problems. These examples were what gave me the tools to solve the homework proof problems.
I found the proposition 1.1.3 very useful in helping undertand the concepts of big-O and little-0.
I am now realizing that I have more to understand about temporal and spatial complexity since I had difficulty explaining part b) of 1.6 and I have my doubts as to whether I understood the implementation during 1.2. I will make sure to ask more question on the subject in the days to come.
Introduction
My Name is Isaac Becker and I am studying applied computational mathematics
This is my fourth year at BYU and I am in the junior core of ACME (applied computational mathematics emphasis)
I have taken all pre-requisites to ACME which include:
Calc 2
Calc 3
Linear Algebra
Differential equations
Math 290
Theory of Analysis
I chose ACME because I want to have a foundation in mathematics before I go into the field of engineering. I am most passionate about modeling and I want to have the competitive skills that ACME is providing.
The professors that provide examples and time during class to have a final answer have had the biggest influence in my learning.
Something interesting about me is that I was on a rowing team in highschool
Plenty of TA hours and the professor's office hours work for me!
Subscribe to:
Posts (Atom)