I'm a first year computer science student, having returned several years after finishing high school.
As I delve deeper down the rabbit hole I am gain a greater appreciation for the importance of math in computer science.
The problem is I don't enjoy math, I read an article called Lockhart's Lament that I think sums up my experience with math in high school. I want to learn more math to improve my programming abilities and broaden my problem solving abilities and at the moment I am working through Rosen's discrete math text but it feels like a grind.
I used to enjoy math as a kid, and I naturally enjoy problem solving and logic. I am hoping to somehow rekindle my interest in math. Has anyone managed to rekindle their interest in math for its own sake?
Where did you start, what type of things did you do? I'd be interested to hear other peoples experiences, maybe inspiring books or good text books
The problem your solving with the math as you learn it must to be important to you, then the math will be worth it. Just reading a book that bores you without working on a problem you enjoy simultaneously probably won't work very well.
If you have been running into programming that requires math, find some variant of that programming that is important to you, and solve it, learning the math required as you go.
I enjoy solving project euler problems a lot. They are mathematical in nature and may sparkle your interest.
The problem with math becoming a grind (from an instructor's perspective) is the notation. Most people read problems or descriptions of concepts and have no trouble. The moment a notation is introduced to describe a problem mathematically, people tend to freak out.
Your question indicates you are specifically interested in discrete mathematics specifically. In a lot of discrete mathematics that I've covered, it is helpful to have a fun puzzle to solve that is specifically designed to focus on the logical and separate elements of a problem at hand.
Attempting puzzles is only one part of it though, because you can get puzzles at any bookstore. If you can't describe the problem mathematically, though, you haven't really learned anything beyond the algorithm in your mind which you may or may not be able to describe programmatically. Solving a cryptogram is a simple dictionary attack that most people can manage through their own experience with language. Trying to program an algorithm to do it logically is another story altogether.
As you find the puzzles you enjoy solving (or attempting to solve), use your textbook(s) to attempt to describe the logical problems involved through appropriate notation. A good example of this would be the many Knights, Knaves, and Normals puzzles on the internet. Get regular logic problems and attempt to write the logical descriptions of each clue/deduction through the notation.
When you get to Big O/Big Theta notation... good luck. I still struggle with this beyond basic understanding of the concepts.
I never enjoyed math either and getting a degree in Engineering was not as simple...I barely managed to get decent grades in high school and college! As I graduated and started working at an IT company, I started gaining interest in finance and accounting. Being weak with numbers made this even more difficult. So, I searched online and came across this YouTube Channel: The Khan Academy. This guy is amazing! He has posted video tutorials on pretty much everything that we learnt in high school, including algebra, probability and arithmetic and much more. His tutorials are easy to understand and a great way to engage with other people who are on the same position as you on the learning curve.
Even better, Bill Gates and his sons follow the tutorials too. Read more about it here: http://money.cnn.com/2010/08/23/technology/sal_khan_academy.fortune/
Related
I recently got interested in Theo Jensen's strandbeest, (If you haven't seen them before look them up! Such incredible engineering.) and I want to mess around with the design he has for his strandbeests' legs. However doing straight forward kinematics is waaay over my head for something like this.
Here's what I'm trying to model:
https://upload.wikimedia.org/wikipedia/commons/6/61/Strandbeest-Walking-Animation.gif
(Can't link directly because I don't have enough reputation :/)
All I really need to know is the path of the 'foot', so something visual isn't necessary.
The final goal is to be able to apply an evolutionary algorithm to it and see if I come up with the same linkage lengths as Theo did, or maybe improve them somehow, so if I there was some software that allowed scripts to be run, that'd be ideal.
Sorry if the question is kind of vague, I'm not all that sure what I'm looking for. Even if there is some maths/engineering topic that would make this easier I'd love to learn.
Thanks!
-Oisin.
Well, I searched for Physics Engine, and found a promising result.
Open Dynamics Engine seems to be an open source physics engine that could fit your needs.
The Open Dynamics Engine (ODE) is a free, industrial quality library for simulating articulated rigid body dynamics. Proven applications include simulating ground vehicles, legged creatures, and moving objects in VR environments. It is fast, flexible and robust, and has built-in collision detection.
Source: Wiki Introduction
There site is ode.org, and it looks like you should be able to evaluate it from there. "[S]imulating rigid body dynamics" is what you want, right? From what I understand, it ought to fit the bill. C++ is probably a reasonable language to attempt this in. I presume you have previous programming experience? This is not what I would consider a beginner's project.
When you get to the evolution, search for Genetic Algorithms. They're frequently used for optimization, and could help you out considerably. Another thing to consider is what you're actually optimizing for (lowest wind speed to function, fasted movement, etc).
I m doing a research project on "Object detection using my a digital camera".
Some suggestion on how to build and program the Matlab code.
In particular, I have a picture of one object, say a screen of my laptop. Than I rotate the laptop and I shot a new picture. I would like to know the difference on the position of the screen. I think I can use the edge detection after a subtraction of the two images but... it is quite difficult for me to implement it.
Some suggestion on how to build and program the matlab code.
That largely depends on the goal you want to achieve. Can you be more specific? Are you streaming the frames or are you tracking offline?
In particular, i have a picture of one object, say a screen of my laptop. Than i rotate the laptop and i shot a new picture. I would like to know the difference on the position of the screen.
There are many ways to do this, and an extensive litterature on the subject. I don't believe anyone would write up the equivalent of a survey paper on the subject as an answer on StackOverflow. Why don't you get started with an object tracking survey paper and then ask a more precise question?
hi, I m doing a reasearch project on "Object detection using my a digital camera". [...] I think i can use the edge detection after a subtraction of the two images but...is quite difficult for me to implement it.
What is your question? Are you asking us if this is a good way to track objects? Are you asking us if this is a new approach and has never been done? Are you asking someone to implement it for you?
Object tracking is a hard problem. I doubt that technique would succeed in any but the most basic scenarios. However, if you look at a survey paper, you might be pointed to a paper that already implemented this an presents results. Finally, I think you should brush up your programming skills because most (successful) object tracking techniques are not trivial to implement. If you don't want to program it yourself, there are online services where you can hire people. StackOverflow is not one of those places.
EDIT: I could deduce that you're new to both programming (in MATLAB) and in object tracking, hence in my answer. Don't mis-understand me, I'm trying to help. Let me re-phrase my suggestions as list:
Your question is far too general. You will get a lot more help from the SO community if you ask more precise questions for two reasons: A) general question result in general answers; and B) the way you asked your question could easily be interpreted as "someone, please do my work for me" even if that's not what you think you're asking.
Get acquainted with the problem domain. To ask more precise questions, you must be close to your answer. For good knowledge on the "object detection and tracking", find a good survey paper. If you're starting off on a research project, people in your lab should be of help to point you to a good one.
Learn to program simple things first. All of the most proficient (effective and efficient) programmers I've ever met struggled with the bubble sort when they were introduced to sorting. None of them would have been able to program an object detection algorithm as a first assignment. Get yourself a good image processing book that has exercises in MATLAB, go through execises one by one. If you can't do them all, choose those that are relevant to what you're trying to accomplish.
I'm at a fork in the road. I need somebody to give me some advice from their personal journey in IT.
At the moment, I have a college diploma (2 years) in Computer Programmer, and about 2 years of professional experience in the field of software. I'm currently freelancing my programming skills to the public, and am enjoying a nice income, and the rewards of flexibly working on a variety of projects with different cool people.
I'm young (21 years old), passionate about software, technology, the internet, and also business. I know if I ever want to dwell deeper into the software industry, I might have a hard time doing so without a Bachelors in Computer Science.
On one side, I think I'm better off getting my BCS while I'm still young and malleable. Also, the thought of learning even more stuff in my field is really exciting to me. On the flip side, it means another 3-4 years of studying, and jeopardizing my chances of going on vacation and accumulating wealth for a long time.
Considering that I'm already pretty successful with my college diploma, do you think it's a good idea for me to go get my BCS? Will it open up many more doors in the future?
I know if I ever want to dwell deeper into the software industry, I might have a hard time doing so without a Bachelors in Computer Science.
I completely disagree with this statement. If you want to "dwell deeper into the software industry", your drive, determination, and will to learn will be your limiting factors. I've known many CS Masters Degree candidates who, frankly, sucked. I've known CS majors who were very skilled and talented. I've known English majors who were excellent developers. The distinguishing factor between those that could and those that could not was a desire to learn.
I once thought that I needed a degree, but ultimately found it to be a distraction. I even wrote about it here: http://jasonleveille.com/blog/2009/10/the-cs-masters-degree-distraction. I don't know if a degree is or isn't right for you. You have to make that decision. Good luck figuring it out.
As far as helping your resume for getting a good job... I consider my stackoverflow profile about as powerful as my education background from one of the top universities in Canada. Just a thought.
I do think a diploma will help you though in 2 main ways:
In learning ability/thought discipline
As Resume flair for getting jobs
But experience is king for many companies. Personally if I was in your shoes I'd probably continue to do some contract work and take a class or two at a time.
If you have the will now though, and the money, and the time, go for the education full time and take on some projects part time. You may not be as flexible in the future.
My journey is a bit different from yours as I did my Bachelor's straight out of high school. I graduated with a high enough average to get a few scholarships and ended up with CS as one of my majors and Combinatorics & Optimization as my other one. I graduated in 1997 and this was about the time of the dot-com boom so while it did take a few months I did eventually get a job.
Could some of your college credits count towards a Bachelor's degree? Have you considered seeing if there is a Bachelor of Information Systems that may suit you better than a general CS program? Those are a couple of questions that I'd put out there as I'd likely think the key is finding that program that works for you is the key point I'd have as while some places may have that Bachelor's degree as a requirement, there will likely be places that wouldn't have that as a requirement so I kind of second Jason's reply on that point.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
Questions asking us to recommend or find a tool, library or favorite off-site resource are off-topic for Stack Overflow as they tend to attract opinionated answers and spam. Instead, describe the problem and what has been done so far to solve it.
Closed 8 years ago.
Improve this question
I'm really interested in Artificial Neural Networks, but I'm looking for a place to start.
What resources are out there and what is a good starting project?
First of all, give up any notions that artificial neural networks have anything to do with the brain but for a passing similarity to networks of biological neurons. Learning biology won't help you effectively apply neural networks; learning linear algebra, calculus, and probability theory will. You should at the very least make yourself familiar with the idea of basic differentiation of functions, the chain rule, partial derivatives (the gradient, the Jacobian and the Hessian), and understanding matrix multiplication and diagonalization.
Really what you are doing when you train a network is optimizing a large, multidimensional function (minimizing your error measure with respect to each of the weights in the network), and so an investigation of techniques for nonlinear numerical optimization may prove instructive. This is a widely studied problem with a large base of literature outside of neural networks, and there are plenty of lecture notes in numerical optimization available on the web. To start, most people use simple gradient descent, but this can be much slower and less effective than more nuanced methods like
Once you've got the basic ideas down you can start to experiment with different "squashing" functions in your hidden layer, adding various kinds of regularization, and various tweaks to make learning go faster. See this paper for a comprehensive list of "best practices".
One of the best books on the subject is Chris Bishop's Neural Networks for Pattern Recognition. It's fairly old by this stage but is still an excellent resource, and you can often find used copies online for about $30. The neural network chapter in his newer book, Pattern Recognition and Machine Learning, is also quite comprehensive. For a particularly good implementation-centric tutorial, see this one on CodeProject.com which implements a clever sort of network called a convolutional network, which constrains connectivity in such a way as to make it very good at learning to classify visual patterns.
Support vector machines and other kernel methods have become quite popular because you can apply them without knowing what the hell you're doing and often get acceptable results. Neural networks, on the other hand, are huge optimization problems which require careful tuning, although they're still preferable for lots of problems, particularly large scale problems in domains like computer vision.
I'd highly recommend this excellent series by Anoop Madhusudanan on Code Project.
He takes you through the fundamentals to understanding how they work in an easy to understand way and shows you how to use his brainnet library to create your own.
Here are some example of Neural Net programming.
http://www.codeproject.com/KB/recipes/neural_dot_net.aspx
you can start reading here:
http://web.archive.org/web/20071025010456/http://www.geocities.com/CapeCanaveral/Lab/3765/neural.html
I for my part have visited a course about it and worked through some literature.
Neural Networks are kind of declasse these days. Support vector machines and kernel methods are better for more classes of problems then backpropagation. Neural networks and genetic algorithms capture the imagination of people who don't know much about modern machine learning but they are not state of the art.
If you want to learn more about AI and machine learning, I recommend reading Peter Norvig's Artificial Intelligence: A Modern Approach. It's a broad survey of AI and lots of modern technology. It goes over the history and older techniques too, and will give you a more complete grounding in the basics of AI and machine Learning.
Neural networks are pretty easy, though. Especially if you use a genetic algorithm to determine the weights, rather then proper backpropagation.
I second dwf's recommendation of Neural Networks for Pattern Recognition by Chris Bishop. Although, it's perhaps not a starter text. Norvig or an online tutorial (with code in Matlab!) would probably be a gentler introduction.
A good starter project would be OCR (Optical Character Recognition). You can scan in pages of text and feed each character through the network in order to perform classification. (You would have to train the network first of course!).
Raul Rojas' book is a a very good start (it's also free). Also, Haykin's book 3rd edition, although of large volume, is very well explained.
I can recommend where not to start. I bought An Introduction to Neural Networks by Kevin Gurney which has good reviews on Amazon and claims to be a "highly accessible introduction to one of the most important topics in cognitive and computer science". Personally, I would not recommend this book as a start. I can comprehend only about 10% of it, but maybe it's just me (English is not my native language). I'm going to look into other options from this thread.
http://www.ai-junkie.com/ann/evolved/nnt1.html is a clear introduction to multi-layers perceptron, although it does not describe the backpropagation algorithm
you can also have a look at generation5.org which provides a lot of articles about AI in general and has some great texts about neural network
If you don't mind spending money, The Handbook of Brain Theory and Neural Networks is very good. It contains 287 articles covering research in many disciplines. It starts with an introduction and theory and then highlights paths through the articles to best cover your interests.
As for a first project, Kohonen maps are interesting for categorization: find hidden relationships in your music collection, build a smart robot, or solve the Netflix prize.
I think a good starting point would always be Wikipedia. There you'll find some usefull links to documentations and projects which use neural nets, too.
Two books that where used during my study:
Introductional course: An introduction to Neural Computing by Igor Aleksander and Helen Morton.
Advanced course: Neurocomputing by Robert Hecht-Nielsen
I found Fausett's Fundamentals of Neural Networks a straightforward and easy-to-get-into introductory textbook.
I found the textbook "Computational Intelligence" to be incredibly helpful.
Programming Collective Intelligence discusses this in the context of Search and Ranking algorithms. Also, in the code available here (in ch.4), the concepts discussed in the book are illustrated in a Python example.
I agree with the other people who said that studying biology is not a good starting point... because theres a lot of irrelevant info in biology. You do not need to understand how a neuron works to recreate its functionality - you only need to simulate its actions. I recomend "How To Create A Mind" by Ray Kurzweil - it goes into the aspect of biology that is relevant for computational models, (creating a simualted neuron by combining several inputs and firing once a threshhold is reached) but ignores the irrelvant stuff like how the neuron actually adds thouse inputs togeather. (You will just use + and an inequality to compare to a threshold, for example)
I should also point out that the book isn't really about 'creating a mind' - it only focuses on heirarchical pattern recognition / the neocortex. The general theme has been talked about since the 1980s I beleive, so there are plenty of older books that probably contain slightly dated forms of the same information. I have read older documents stating that the vision system, for example, is a multi layered pattern recognizer. He contends that this applies to the entire neocortex. Also, take his 'predictions' with a grain of salt - his hardware estimates are probably pretty accurate, but i think he underestimates how complicated simple tasks can be (ex: driving a car). Granted, he has seen a lot of progress (and been part of some of it) but i still think he is over optimistic. There is a big difference between an AI car being able to drive a mile successfully 90% of the time, when compared to the 99.9+% that a human can do. I don't expect any AI to be truly out driving me for atleast 20 years... (I don't count BMWs track cars that need to be 'trained' on the actual course, as they aren't really playing the same game)
If you already have a basic idea of what AI is and how it can be modeled, you may be better off skipping to something more technical.
If you want to do quickly learn about applications of some neural network concepts on a real simulator, there is a great online book (now wiki) called 'Computational Cognitive Neuroscience' at http://grey.colorado.edu/CompCogNeuro/index.php/CCNBook/Main
The book is used at schools as a textbook, and takes you through lots of different brain areas, from individual neurons all the way to higher-order executive functioning.
In addition, each section is augmented with homework 'projects' that are already down for you. Just download, follow the steps, and simulate everything that the chapter talked about. The software they use, Emergent, is a little finnicky but incredibly robust: its the product of more than 10 years of work I believe.
I went through it in an undergrad class this past semester, and it was great. Walks you through everything step by step
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 9 years ago.
Improve this question
We don't teach children calculus first. We first teach them arithmetic, then algebra, then geometry, the analytical geometry, then finally calculus.
Why then, do we teach our computer scientists frameworks and IDE first. Some curriculum do force students to learn computer science fundamentals, but the vast majority of graduates that I see could not compose a framework of their own to save their lives.
Where then is the next generation of tool builders?
How can we promote the understanding necessary to create frameworks and development environments?
This is of course a generality. Not all education is lacking, but it seems to be the majority and it brings down the quality of our profession as a whole.
I think the analogy is a bit off. A better analogy would be "We don't teach our kids to use calculators to add and subtract, why teach programmers to use an IDE to program?"
Get rid of HR departments that require X years experience in Y. The universities are just tailoring their course to the HR department's requirements.
I employ graduates who can code in something (I really don't care what language) and who can learn.
I see your point, although I think the math analogy doesn't quite fit. You have to know basic arithmetic to be able to get anything done in any other math discipline.
When I began programming frameworks were mostly unheard of. If you wanted a binary tree, by God, you went and wrote one. In C or Assembler. That was basically it, so to get anything done at all you had to know a lot.
Today, Frameworks and IDEs and designers make it possible for "noobs" to create actually pretty brilliant things without knowing the first thing about how to build a framework, or a compiler, or manage memory allocation.
The real issue is, what about all the dingbats that think they are awesome, great programmers because they used Frontpage or Access? Managers have a hard time telling the difference between that kind of programmer and one that really knows software development as a discipline.
So, specifically, why is it that way? Because everyone wants a job and nobody hires programmers that know how to build a binary tree. They want programmers that know .Net or J2EE, etc.
I would argue that there is probably enough work out there for 9 to 5 programmers who can start at the framework level and go up from there. The truly good ones - mostly your program as a career and/or program as a hobby - are going to get the knowledge they may have missed in college over time anyway. You can't force everyone to be a wonderful programmer no matter what curriculum you teach. Inquisitive students are going to learn about the fundamentals whether its taught to them in class or entirely on their own.
There are tool makers and tool breakers. And of course there are tools, but let's not go there.
If you have a good look at an automotive workshop, you will see a lot of funny little tools that you don't see on the shelves in hardware stores. Like the ones for pushing back brake caliper pistons. Or the clamps for compressing valve stems so you can get the collets out with one hand while talking to your mates about nailing the new secretary (instead of watching them fly across the room when the spring slips out from your screwdriver).
These were designed by mechanics. They're really effective, generally small and cheap, and totally incomprehensible until you seen them in action.
Most of the profound changes in automotive technology were bottom-up, but top-down is also needed. Individual mechanics can't make fundamental technology changes like the switch from cast iron to alloy heads. A new broom sweeps clean, an old broom knows the corners. You need both.
But I digress: the point is that the mechanics couldn't design these tools if they lacked fundamental skills and knowledge. My father built me an entire motorcycle from scrap iron when I was a kid. As an adult, because I lack his skills and knowledge and modes of thought, I can barely maintain the bike I bought from Honda, much less take to it with an oxy like Mr T in a creative frenzy.
With code, I am as my father was with steel. Donald Knuth is my constant companion, and when the wireless protocol for our GPS loggers needs to be implemented in .NET it's me they come to see. The widget monkeys wouldn't know where to start.
I think the problem is in fact the GUI paradigm in general.
Microsoft made using computers much easier, they popularized the Graphical User Interface. They brought this interface metaphor, (the desktop, the file) to the domain of programming as well and very effectively too with their Visual Basic tool.
But just as the GUI obscures what happens "under the hood" so does the IDE obscure the manipulation of bits and bytes. The question is, of course, risk to reward ratio - how much understanding do programmers lose in exchange for productivity?
A cursory look at "The Art of Computer Programming" might show why IDEs are useful; "The ultimate packing density is achieved when we have 1-bit items, because we can cram 64 of them into a single 64-bit word. Suppose, for example, that we want a table of all odd prime numbers less than 1024, so that we can easily decide the primality of a small integer. No problem; only eight 64-bit numbers are required:
p0 = 011101101101001100101101001001001100101100101001000101101101000000
p1 = . . ."
Programming is really hard, you can see how an IDE might help. :^)
Learning the abstraction is easier than learning the details when it comes to programming. It's harder to teach someone to hand-code assembler to print "Hello World" than it is to have them throw together a form with a button on it that shows a "Hello World" message when the button is clicked.
You didn't know how to build the engine of a car before learning to drive, did you? Because it's not necessary in order to drive. In the same vein, you don't need to learn how a linked list or binary tree works in order to maintain a list of names and search them.
There will always be those who want to get under the hood and learn the "why" of things, but I don't think it's required to get things done.
I always screen applications by asking difficult questions that they could only answer if they understood how something really works. I think it is a real shame colleges and universities are teaching people framework based development but not focusing on core software principles. I agree that what matters more than anything is someone who understands how programming works and has the drive to learn anything they can about it.
Most universities I know of have an introduction to computer programming course that teaches basic programming concepts. Unfortunately it is impossible to teach programming without actually writing code.
The problem is that some prefer to teach this course using some OO language such as JAVA or C# and so the students must use Visual Studio (or the Java equivalent).
It is very hard to explain the basic concepts when the IDE forces you to work in a certain way.
I think that the first language students learn should be functional language such as C. This way you have less layers of abstraction between them and the basic CS concepts.
Agree with cfeduke.
I looked at the work for the same CS courses I did from 2 years previously, and they were way harder. 5 years previously, way way harder.
The CS bar is being lowered more and more, presumably because there are more and more jobs that don't require any working knowledge of any of the complicated CS subjects. There are huge numbers of jobs for people to just cut code.
Since traditionaly people who wanted to be programmers did CS courses as coding has gotten easier this is still the case.
What really needs to happen is for CS to not be a requirement for professional software development. Instead there needs to be another curriculam that focuses more on getting people out the door and cutting code.
This would leave CS to be that course for you next generation of tool builder.