In ML videos of Andrew Ng on Coursera on Classification (in the third video), he said that the "decision boundary is not a property of the training set". What does this statement mean? And does it also imply that the straight line or any curves that we use in linear regression to fit data are not a property of the training set? He claims that those curves (achieved through linear regression) aren't the properties of the corresponding training data. I am a bit confused about this. Kindly if my doubts could be removed. Thanks in advance.
The decision boundary is a property of your classifier. Different classifiers lead to different decision boundaries.
Decision boundary has nothing to do with linear regression, as it only makes sense for classification problems. The decision boundary is the curve (or surface, in more than two dimensions) that splits the elements of the two different classes in your classification problem. In logistic regression, the decision boundary is a straight line, while in nonlinear classification methods, like neural networks, the decision boundary is a curve.
Related
I have started to learn image processing and I am trying to learn LDA (Linear Discriminant Analysis) algorithm. I want to ask a question to understand the philosophy of LDA. If it is is useful to use LDA in the distribution in the first example, is it also useful to use LDA in the distribution in the second example. I mean if I rotate 90 degrees to the image, is LDA still useful?
first
second
LDA is a generative algorithm that allows to classify predictors in two or more classes divided by linear boundaries. So, if you rotate the image the LDA algorithm does not lose its value, as long as you fit two different LDA models.
I have some 2d data which looks like it could be well classified by an area described by two intersecting straight lines. These lines won't necessarily by at right angles to each other. Here is a simple example where the two lines would be more or less at right angles:
Is there a suitable classifier for this? Logistic regression will give me one straight line but I am not sure what will give me two as a decision boundary. A decision tree will give me two that are axis parallel which isn't really want I want.
You can give Support Vector Machine (SVM) a try. There are multiple kernels that can be used with SVM, like
Linear
Polynomial
RBF (Radial Basis Function)
Sigmoid
You can even specify custom kernels as mentioned in this list.
Here is an image of decision boundaries over Iris dataset, taken from this example
References
Difference between various SVM kernels
Selecting Kernels for SVM
Custom kernel SVM
I am currently using sklearn's Logistic Regression function to work on a synthetic 2d problem. The dataset is shown as below:
I'm basic plugging the data into sklearn's model, and this is what I'm getting (the light green; disregard the dark green):
The code for this is only two lines; model = LogisticRegression(); model.fit(tr_data,tr_labels). I've checked the plotting function; that's fine as well. I'm using no regularizer (should that affect it?)
It seems really strange to me that the boundaries behave in this way. Intuitively I feel they should be more diagonal, as the data is (mostly) located top-right and bottom-left, and from testing some things out it seems a few stray datapoints are what's causing the boundaries to behave in this manner.
For example here's another dataset and its boundaries
Would anyone know what might be causing this? From my understanding Logistic Regression shouldn't be this sensitive to outliers.
Your model is overfitting the data (The decision regions it found perform indeed better on the training set than the diagonal line you would expect).
The loss is optimal when all the data is classified correctly with probability 1. The distances to the decision boundary enter in the probability computation. The unregularized algorithm can use large weights to make the decision region very sharp, so in your example it finds an optimal solution, where (some of) the outliers are classified correctly.
By a stronger regularization you prevent that and the distances play a bigger role. Try different values for the inverse regularization strength C, e.g.
model = LogisticRegression(C=0.1)
model.fit(tr_data,tr_labels)
Note: the default value C=1.0 corresponds already to a regularized version of logistic regression.
Let us further qualify why logistic regression overfits here: After all, there's just a few outliers, but hundreds of other data points. To see why it helps to note that
logistic loss is kind of a smoothed version of hinge loss (used in SVM).
SVM does not 'care' about samples on the correct side of the margin at all - as long as they do not cross the margin they inflict zero cost. Since logistic regression is a smoothed version of SVM, the far-away samples do inflict a cost but it is negligible compared to the cost inflicted by samples near the decision boundary.
So, unlike e.g. Linear Discriminant Analysis, samples close to the decision boundary have disproportionately more impact on the solution than far-away samples.
I'm trying to understand "Back Propagation" as it is used in Neural Nets that are optimized using Gradient Descent. Reading through the literature it seems to do a few things.
Use random weights to start with and get error values
Perform Gradient Descent on the loss function using these weights to arrive at new weights.
Update the weights with these new weights until the loss function is minimized.
The steps above seem to be the EXACT process to solve for Linear Models (Regression for e.g.)? Andrew Ng's excellent course on Coursera for Machine Learning does exactly that for Linear Regression.
So, I'm trying to understand if BackPropagation does anything more than gradient descent on the loss function.. and if not, why is it only referenced in the case of Neural Nets and why not for GLMs (Generalized Linear Models). They all seem to be doing the same thing- what might I be missing?
The main division happens to be hiding in plain sight: linearity. In fact, extend to question to continuity of the first derivative, and you'll encapsulate most of the difference.
First of all, take note of one basic principle of neural nets (NN): a NN with linear weights and linear dependencies is a GLM. Also, having multiple hidden layers is equivalent to a single hidden layer: it's still linear combinations from input to output.
A "modern' NN has non-linear layers: ReLUs (change negative values to 0), pooling (max, min, or mean of several values), dropouts (randomly remove some values), and other methods destroy our ability to smoothly apply Gradient Descent (GD) to the model. Instead, we take many of the principles and work backward, applying limited corrections layer by layer, all the way back to the weights at layer 1.
Lather, rinse, repeat until convergence.
Does that clear up the problem for you?
You got it!
A typical ReLU is
f(x) = x if x > 0,
0 otherwise
A typical pooling layer reduces the input length and width by a factor of 2; in each 2x2 square, only the maximum value is passed through. Dropout simply kills off random values to make the model retrain those weights from "primary sources". Each of these is a headache for GD, so we have to do it layer by layer.
So, I'm trying to understand if BackPropagation does anything more than gradient descent on the loss function.. and if not, why is it only referenced in the case of Neural Nets
I think (at least originally) back propagation of errors meant less than what you describe: the term "backpropagation of errors" only refered to the method of calculating derivatives of the loss function, instead of e.g. automatic differentiation, symbolic differentiation, or numerical differentiation. No matter what the gradient was then used for (e.g. Gradient Descent, or maybe Levenberg/Marquardt).
They all seem to be doing the same thing- what might I be missing?
They're using different models. If your neural network used linear neurons, it would be equivalent to linear regression.
Can both Naive Bayes and Logistic regression classify both of these dataset perfectly ? My understanding is that Naive Bayes can , and Logistic regression with complex terms can classify these datasets. Please help if I am wrong.
Image of datasets is here:
Lets run both algorithms on two similar datasets to the ones you posted and see what happens...
EDIT The previous answer I posted was incorrect. I forgot to account for the variance in Gaussian Naive Bayes. (The previous solution was for naive bayes using Gaussians with fixed, identity covariance, which gives a linear decision boundary).
It turns out that LR fails at the circular dataset while NB could succeed.
Both methods succeed at the rectangular dataset.
The LR decision boundary is linear while the NB boundary is quadratic (the boundary between two axis-aligned Gaussians with different covariances).
Applying NB the circular dataset gives two means in roughly the same position, but with different variances, leading to a roughly circular decision boundary - as the radius increases, the probability of the higher variance Gaussian increases compared to that of the lower variance Gaussian. In this case, many of the inner points on the inner circle are incorrectly classified.
The two plots below show a gaussian NB solution with fixed variance.
In the plots below, the contours represent probability contours of the NB solution.
This gaussian NB solution also learns the variances of individual parameters, leading to an axis-aligned covariance in the solution.
Naive Bayes/Logistic Regression can get the second (right) of these two pictures, in principle, because there's a linear decision boundary that perfectly separates.
If you used a continuous version of Naive Bayes with class-conditional Normal distributions on the features, you could separate because the variance of the red class is greater than that of the blue, so your decision boundary would be circular. You'd end up with distributions for the two classes which had the same mean (the centre point of the two rings) but where the variance of the features conditioned on the red class would be greater than that of the features conditioned on the blue class, leading to a circular decision boundary somewhere in the margin. This is a non-linear classifier, though.
You could get the same effect with histogram binning of the feature spaces, so long as the histograms' widths were narrow enough. In this case both logistic regression and Naive Bayes will work, based on histogram-like feature vectors.
How would you use Naive Bayes on these data sets?
In the usual form, Naive Bayes needs binary / categorial data.