what is the benefit of using Gradient Descent in the linear regression space? looks like the we can solve the problem (finding theta0-n that minimum the cost func) with analytical method so why we still want to use gradient descent to do the same thing? thanks
When you use the normal equations for solving the cost function analytically you have to compute:
Where X is your matrix of input observations and y your output vector. The problem with this operation is the time complexity of calculating the inverse of a nxn matrix which is O(n^3) and as n increases it can take a very long time to finish.
When n is low (n < 1000 or n < 10000) you can think of normal equations as the better option for calculation theta, however for greater values Gradient Descent is much more faster, so the only reason is the time :)
You should provide more details about yout problem - what exactly are you asking about - are we talking about linear regression in one or many dimensions? Simple or generalized ones?
In general, why do people use the GD?
it is easy to implement
it is very generic optimization technique - even if you change your model to the more general one, you can stil use it
So what about analytical solutions? Well, we do use them, your claim is simply false here (if we are talking in general), for example the OLS method is a closed form, analytical solution, which is widely used. If you can use the analytical solution, it is affordable computationaly (as sometimes GD is simply cheapier or faster) then you can, and even should - use it.
Neverlethles this is always a matter of some pros and cons - analytical solutions are strongly connected to the model, so implementing them can be inefficient if you plan to generalize/change your models in the future. They are sometimes less efficient then their numerical approximations, and sometimes there are simply harder to implement. If none of above is true - you should use the analytical solution, and people do it, really.
To sum up, you rather use GD over analytical solution if:
you are considering changes in the model, generalizations, adding some more complex terms/regularization/modifications
you need a generic method because you do not know much about the future of the code and the model (you are only one of the developers)
analytical solution is more expensive computationaly, and you need efficiency
analytical solution requires more memory, which you do not have
analytical solution is hard to implement and you need easy, simple code
I saw a very good answer from https://stats.stackexchange.com/questions/23128/solving-for-regression-parameters-in-closed-form-vs-gradient-descent
Basically, the reasons are:
1.For most nonlinear regression problems there is no closed form solution.
2.Even in linear regression (one of the few cases where a closed form solution is available), it may be impractical to use the formula. The following example shows one way in which this can happen.
Other reason is that gradient descent is immediately useful when you generalize linear regression, especially if the problem doesn't have a closed-form solution, like for example in Lasso (which adds regularization term consisting on sum of absolute values of weight vector).
Related
I am recently studying Machine Learning with Coursera ML course, and some questions popped up while learning cost function with regularization.
Please give me your advice if you have any idea.
If I have enough number of training data, I think regularization would reduce the accuracy because the model is able to obtain high reliability and generalized output only from the training set, without regularization. How can I make a good decision whether or not I should use regularization?
Let’s suppose we have a model as follows: w3*x3 + w2*x2 + w1*x1 +w0, and x3 is the feature which particularly causes overfitting; this means it has more outliers. In this situation, I think the way of regularization is sort of unreasonable due to the fact that it takes effect on every weight. Do you know any better way that I can use in this case?
What is the best way to choose the value of lambda? I guess the simplest way is to conduct multiple learning with different lambda values and to compare their training accuracy. However, this is definitely inefficient when we have huge number of training data. I want to know how you choose the ideal lambda value.
Thanks for reading!
It's a bad idea to come up with guesses before you evaluate your model on validation data. When you talk about 'accuracy' in your question, to which accuracy do you refer to? Train set accuracy is not very useful in estimation of your model's goodness. Normally, regularization is desirable for many families of ML algorithms. In the case of linear regression, it is definitely worth to do. The question here is only the amount of it, i.e. the value of lambda parameter. Also, you might want to try L1 instead of L2. Read this.
In machine learning, questions like this are normally answered using data. Try a model, investigate how it behaves, try different solutions for the issues you observe.
Read this: How to calculate the regularization parameter in linear regression
I have read this article about autoencoder, which is introduced by Andrew Ng. In there, he use sparity like regularization to drop connection but formular of sparsity is different from regur. So, I want to know why we don't use directly regularization term like model NNs or logistic regression : (1/2 * m) * Theta^2 ?
First, let us start with some naming convention, both sparsity penalty and L2 penalty on weights can (and often are) called regularizers. Thus, the question should be "why use sparsity-based regularization instead of simple L2-norm based?". And there is no simple answer for this problem, since it goes not deeply into underlying mathematics and asks what is a better way to make sure our network creates a well generalizing representation - to keep parameters more or less in fixed sphere (L2 regularization, the one you propose) or to make sure that whatever we put as an input to the network, it will produce relatively simple representation (possibly at the cost of having lots of weights/neurons that rarely are used). Even on this level of abstraction it should show qualitatitative difference between these two regularizers, which will lead to building completely differnet models. Will sparsity term be better always? Probably not, nearly nothing in ML is "always better". But on average it seems like a less heuristic choice for an autoencoder - you want to have a kind of compression - thus you force your net to create compressed representation which is really ... well.. compressed (small!), while using L2 regularization would simply "squash" representation in terms of norm (since dot product through weights with small norm will not increase too much norm of the input), but it can still use "tiny bit" of each neuron, thus efficiently build a complex representation (using many units) but simply - with small activations.
When we have a high degree linear polynomial that is used to fit a set of points in a linear regression setup, to prevent overfitting, we use regularization, and we include a lambda parameter in the cost function. This lambda is then used to update the theta parameters in the gradient descent algorithm.
My question is how do we calculate this lambda regularization parameter?
The regularization parameter (lambda) is an input to your model so what you probably want to know is how do you select the value of lambda. The regularization parameter reduces overfitting, which reduces the variance of your estimated regression parameters; however, it does this at the expense of adding bias to your estimate. Increasing lambda results in less overfitting but also greater bias. So the real question is "How much bias are you willing to tolerate in your estimate?"
One approach you can take is to randomly subsample your data a number of times and look at the variation in your estimate. Then repeat the process for a slightly larger value of lambda to see how it affects the variability of your estimate. Keep in mind that whatever value of lambda you decide is appropriate for your subsampled data, you can likely use a smaller value to achieve comparable regularization on the full data set.
CLOSED FORM (TIKHONOV) VERSUS GRADIENT DESCENT
Hi! nice explanations for the intuitive and top-notch mathematical approaches there. I just wanted to add some specificities that, where not "problem-solving", may definitely help to speed up and give some consistency to the process of finding a good regularization hyperparameter.
I assume that you are talking about the L2 (a.k. "weight decay") regularization, linearly weighted by the lambda term, and that you are optimizing the weights of your model either with the closed-form Tikhonov equation (highly recommended for low-dimensional linear regression models), or with some variant of gradient descent with backpropagation. And that in this context, you want to choose the value for lambda that provides best generalization ability.
CLOSED FORM (TIKHONOV)
If you are able to go the Tikhonov way with your model (Andrew Ng says under 10k dimensions, but this suggestion is at least 5 years old) Wikipedia - determination of the Tikhonov factor offers an interesting closed-form solution, which has been proven to provide the optimal value. But this solution probably raises some kind of implementation issues (time complexity/numerical stability) I'm not aware of, because there is no mainstream algorithm to perform it. This 2016 paper looks very promising though and may be worth a try if you really have to optimize your linear model to its best.
For a quicker prototype implementation, this 2015 Python package seems to deal with it iteratively, you could let it optimize and then extract the final value for the lambda:
In this new innovative method, we have derived an iterative approach to solving the general Tikhonov regularization problem, which converges to the noiseless solution, does not depend strongly on the choice of lambda, and yet still avoids the inversion problem.
And from the GitHub README of the project:
InverseProblem.invert(A, be, k, l) #this will invert your A matrix, where be is noisy be, k is the no. of iterations, and lambda is your dampening effect (best set to 1)
GRADIENT DESCENT
All links of this part are from Michael Nielsen's amazing online book "Neural Networks and Deep Learning", recommended reading!
For this approach it seems to be even less to be said: the cost function is usually non-convex, the optimization is performed numerically and the performance of the model is measured by some form of cross validation (see Overfitting and Regularization and why does regularization help reduce overfitting if you haven't had enough of that). But even when cross-validating, Nielsen suggests something: you may want to take a look at this detailed explanation on how does the L2 regularization provide a weight decaying effect, but the summary is that it is inversely proportional to the number of samples n, so when calculating the gradient descent equation with the L2 term,
just use backpropagation, as usual, and then add (λ/n)*w to the partial derivative of all the weight terms.
And his conclusion is that, when wanting a similar regularization effect with a different number of samples, lambda has to be changed proportionally:
we need to modify the regularization parameter. The reason is because the size n of the training set has changed from n=1000 to n=50000, and this changes the weight decay factor 1−learning_rate*(λ/n). If we continued to use λ=0.1 that would mean much less weight decay, and thus much less of a regularization effect. We compensate by changing to λ=5.0.
This is only useful when applying the same model to different amounts of the same data, but I think it opens up the door for some intuition on how it should work, and, more importantly, speed up the hyperparametrization process by allowing you to finetune lambda in smaller subsets and then scale up.
For choosing the exact values, he suggests in his conclusions on how to choose a neural network's hyperparameters the purely empirical approach: start with 1 and then progressively multiply÷ by 10 until you find the proper order of magnitude, and then do a local search within that region. In the comments of this SE related question, the user Brian Borchers suggests also a very well known method that may be useful for that local search:
Take small subsets of the training and validation sets (to be able to make many of them in a reasonable amount of time)
Starting with λ=0 and increasing by small amounts within some region, perform a quick training&validation of the model and plot both loss functions
You will observe three things:
The CV loss function will be consistently higher than the training one, since your model is optimized for the training data exclusively (EDIT: After some time I've seen a MNIST case where adding L2 helped the CV loss decrease faster than the training one until convergence. Probably due to the ridiculous consistency of the data and a suboptimal hyperparametrization though).
The training loss function will have its minimum for λ=0, and then increase with the regularization, since preventing the model from optimally fitting the training data is exactly what regularization does.
The CV loss function will start high at λ=0, then decrease, and then start increasing again at some point (EDIT: this assuming that the setup is able to overfit for λ=0, i.e. the model has enough power and no other regularization means are heavily applied).
The optimal value for λ will be probably somewhere around the minimum of the CV loss function, it also may depend a little on how does the training loss function look like. See the picture for a possible (but not the only one) representation of this: instead of "model complexity" you should interpret the x axis as λ being zero at the right and increasing towards the left.
Hope this helps! Cheers,
Andres
The cross validation described above is a method used often in Machine Learning. However, choosing a reliable and safe regularization parameter is still a very hot topic of research in mathematics.
If you need some ideas (and have access to a decent university library) you can have a look at this paper:
http://www.sciencedirect.com/science/article/pii/S0378475411000607
I'm pretty new in the field of machine learning (even if I find it extremely interesting), and I wanted to start a small project where I'd be able to apply some stuff.
Let's say I have a dataset of persons, where each person has N different attributes (only discrete values, each attribute can be pretty much anything).
I want to find clusters of people who exhibit the same behavior, i.e. who have a similar pattern in their attributes ("look-alikes").
How would you go about this? Any thoughts to get me started?
I was thinking about using PCA since we can have an arbitrary number of dimensions, that could be useful to reduce it. K-Means? I'm not sure in this case. Any ideas on what would be most adapted to this situation?
I do know how to code all those algorithms, but I'm truly missing some real world experience to know what to apply in which case.
K-means using the n-dimensional attribute vectors is a reasonable way to get started. You may want to play with your distance metric to see how it affects the results.
The first step to pretty much any clustering algorithm is to find a suitable distance function. Many algorithms such as DBSCAN can be parameterized with this distance function then (at least in a decent implementation. Some of course only support Euclidean distance ...).
So start with considering how to measure object similarity!
In my opinion you should also try expectation-maximization algorithm (also called EM). On the other hand, you must be careful while using PCA because this algorithm may reduce the dimensions relevant to clustering.
How should I approach a situtation when I try to apply some ML algorithm (classification, to be more specific, SVM in particular) over some high dimensional input, and the results I get are not quite satisfactory?
1, 2 or 3 dimensional data can be visualized, along with the algorithm's results, so you can get the hang of what's going on, and have some idea how to aproach the problem. Once the data is over 3 dimensions, other than intuitively playing around with the parameters I am not really sure how to attack it?
What do you do to the data? My answer: nothing. SVMs are designed to handle high-dimensional data. I'm working on a research problem right now that involves supervised classification using SVMs. Along with finding sources on the Internet, I did my own experiments on the impact of dimensionality reduction prior to classification. Preprocessing the features using PCA/LDA did not significantly increase classification accuracy of the SVM.
To me, this totally makes sense from the way SVMs work. Let x be an m-dimensional feature vector. Let y = Ax where y is in R^n and x is in R^m for n < m, i.e., y is x projected onto a space of lower dimension. If the classes Y1 and Y2 are linearly separable in R^n, then the corresponding classes X1 and X2 are linearly separable in R^m. Therefore, the original subspaces should be "at least" as separable as their projections onto lower dimensions, i.e., PCA should not help, in theory.
Here is one discussion that debates the use of PCA before SVM: link
What you can do is change your SVM parameters. For example, with libsvm link, the parameters C and gamma are crucially important to classification success. The libsvm faq, particularly this entry link, contains more helpful tips. Among them:
Scale your features before classification.
Try to obtain balanced classes. If impossible, then penalize one class more than the other. See more references on SVM imbalance.
Check the SVM parameters. Try many combinations to arrive at the best one.
Use the RBF kernel first. It almost always works best (computationally speaking).
Almost forgot... before testing, cross validate!
EDIT: Let me just add this "data point." I recently did another large-scale experiment using the SVM with PCA preprocessing on four exclusive data sets. PCA did not improve the classification results for any choice of reduced dimensionality. The original data with simple diagonal scaling (for each feature, subtract mean and divide by standard deviation) performed better. I'm not making any broad conclusion -- just sharing this one experiment. Maybe on different data, PCA can help.
Some suggestions:
Project data (just for visualization) to a lower-dimensional space (using PCA or MDS or whatever makes sense for your data)
Try to understand why learning fails. Do you think it overfits? Do you think you have enough data? Is it possible there isn't enough information in your features to solve the task you are trying to solve? There are ways to answer each of these questions without visualizing the data.
Also, if you tell us what the task is and what your SVM output is, there may be more specific suggestions people could make.
You can try reducing the dimensionality of the problem by PCA or the similar technique. Beware that PCA has two important points. (1) It assumes that the data it is applied to is normally distributed and (2) the resulting data looses its natural meaning (resulting in a blackbox). If you can live with that, try it.
Another option is to try several parameter selection algorithms. Since SVM's were already mentioned here, you might try the approach of Chang and Li (Feature Ranking Using Linear SVM) in which they used linear SVM to pre-select "interesting features" and then used RBF - based SVM on the selected features. If you are familiar with Orange, a python data mining library, you will be able to code this method in less than an hour. Note that this is a greedy approach which, due to its "greediness" might fail in cases where the input variables are highly correlated. In that case, and if you cannot solve this problem with PCA (see above), you might want to go to heuristic methods, which try to select best possible combinations of predictors. The main pitfall of this kind of approaches is the high potential of overfitting. Make sure you have a bunch "virgin" data that was not seen during the entire process of model building. Test your model on that data only once, after you are sure that the model is ready. If you fail, don't use this data once more to validate another model, you will have to find a new data set. Otherwise you won't be sure that you didn't overfit once more.
List of selected papers on parameter selection:
Feature selection for high-dimensional genomic microarray data
Oh, and one more thing about SVM. SVM is a black box. You better figure out what is the mechanism that generate the data and model the mechanism and not the data. On the other hand, if this would be possible, most probably you wouldn't be here asking this question (and I wouldn't be so bitter about overfitting).
List of selected papers on parameter selection
Feature selection for high-dimensional genomic microarray data
Wrappers for feature subset selection
Parameter selection in particle swarm optimization
I worked in the laboratory that developed this Stochastic method to determine, in silico, the drug like character of molecules
I would approach the problem as follows:
What do you mean by "the results I get are not quite satisfactory"?
If the classification rate on the training data is unsatisfactory, it implies that either
You have outliers in your training data (data that is misclassified). In this case you can try algorithms such as RANSAC to deal with it.
Your model(SVM in this case) is not well suited for this problem. This can be diagnozed by trying other models (adaboost etc.) or adding more parameters to your current model.
The representation of the data is not well suited for your classification task. In this case preprocessing the data with feature selection or dimensionality reduction techniques would help
If the classification rate on the test data is unsatisfactory, it implies that your model overfits the data:
Either your model is too complex(too many parameters) and it needs to be constrained further,
Or you trained it on a training set which is too small and you need more data
Of course it may be a mixture of the above elements. These are all "blind" methods to attack the problem. In order to gain more insight into the problem you may use visualization methods by projecting the data into lower dimensions or look for models which are suited better to the problem domain as you understand it (for example if you know the data is normally distributed you can use GMMs to model the data ...)
If I'm not wrong, you are trying to see which parameters to the SVM gives you the best result. Your problem is model/curve fitting.
I worked on a similar problem couple of years ago. There are tons of libraries and algos to do the same. I used Newton-Raphson's algorithm and a variation of genetic algorithm to fit the curve.
Generate/guess/get the result you are hoping for, through real world experiment (or if you are doing simple classification, just do it yourself). Compare this with the output of your SVM. The algos I mentioned earlier reiterates this process till the result of your model(SVM in this case) somewhat matches the expected values (note that this process would take some time based your problem/data size.. it took about 2 months for me on a 140 node beowulf cluster).
If you choose to go with Newton-Raphson's, this might be a good place to start.