I am a beginner with tensorflow and I want to implement MLP and train it based on the back propagation algorithm but when I read tutorials I found that it uses optmizers like “Stochastic Gradient Descent” and called it back propagation without implementing the algorithm phases. How is this back propagation?
Typically you use these optimizers by first defining the loss function and then invoking optimizer's .optimize on it:
loss = some_loss(predicted_outputs, true_outputs)
optimizer = tf.train.GradientDescentOptimizer(learning_rate)
train_op = optimizer.minimize(loss)
Now, executing sess.run(train_op) will make a step in the direction of the gradients of weights to minimize loss.
I think this may help：
In [18], two approaches for back-propagating gradients through a
computational graph are described. The first, which the authors refer
to as symbol-to-number differentiation, re- ceives a set of input
values and then computes the numerical values of the gradients at
those input values. It does so by explicitly traversing the graph
first in the forward order (forward-propagation) to compute the cost,
then in reverse order (back-propagation) to compute the gradients via
the chain rule. Another approach, more relevant to TensorFlow, is what
[18] calls symbol-to-symbol derivatives and [8] terms automatic
gradient computation. In this case, gradients are not computed by an
explicit implementation of the back- propagation algorithm. Rather,
special nodes are added to the computational graph that calculate the
gradient of each operation and thus ultimately the chain rule. To
perform back- propagation, these nodes must then simply be executed
like any other nodes by the graph evaluation engine. As such, this
approach does not produce the desired derivatives as a numeric value,
but only as a symbolic handle to compute those values.
Reference: http://arxiv.org/abs/1610.0117
Related
When training a neural network, if the same module is used multiple times in one iteration, does the gradient of the module need special processing during backpropagation?
for example:
One Deformable Compensation is used three times in this model, which means they share the same weights.
What will happen when I use loss.backward()?
Will loss.backward() work correctly？
The nice thing about autograd and backward passes is that the underlying framework is not "algorithmic", but rather a mathematic one: it implements the chain rule of derivatives. Therefore, there are no "algorithmic" considerations of "shared weights" or "weighting different layers", it's pure math. The backward pass provides the derivative of your loss function w.r.t the weights in a purely mathematical way.
Sharing weights can be done globally (e.g., when training Saimese networks), on a "layer level" (as in your example), but also within a layer. When you think about it Convolution layers and Reccurent layers are a fancy way of locally sharing weights.
Naturally, pytorch (as well as all other DL frameworks) can trivially handle these cases.
As long as your "deformable compensation" layer is correctly implemented -- pytorch will take care of the gradients for you, in a mathematically correct manner, thanks to the chain rule.
I am trying to design a simple a neural network but I am facing some problems.
My output keeps converging to 0.5 when I use ReLU and to 0.7 when I use sigmoid function.
Please suggest me :
Is there any other way of initializing weights and biases ? I am initializing it randomly.
Do I need to perform backprop after every forward pass or I should take average op errors and update it at the end of the epoch ?
Do I need to use biases in the input layer ?
I am new to neural networks. Please help.
Is there any other way of initializing weights and biases ? I am initializing it randomly.
Yes, it is common to initialize the weights randomly. However, there are different techniques for choosing the variance, e.g. Xavier initialization, He initialization, etc (see this discussion).
It is a bit different for the biases. Unlike weights, it's totally ok to init them with zeros. In ReLu based networks, it's common to use slightly positive numbers to ensure that most activations are positive, at least initially and backpropagate the error. But random init usually works as well.
Do I need to perform backprop after every forward pass or I should take average op errors and update it at the end of the epoch ?
In a classical algorithm, yes. The idea is to assess and update the network iteratively. But you can perform both operations for a mini-batch, instead of individual instances, if that's what you describe, and it works more efficiently. But it's not common to do several forward passes before one backward pass, it will only slow down the training.
Do I need to use biases in the input layer ?
The biases appear in the linear layer, along with the weights. The data itself is passed to the first layer without a bias.
I am trying to simulate a XOR gate using a neural network similar to this:
Now I understand that each neuron has certain number of weights and a bias. I am using a sigmoid function to determine whether a neuron should fire or not in each state (since this uses a sigmoid rather than a step function, I use firing in a loose sense as it actually spits out real values).
I successfully ran the simulation for feed-forwarding part, and now I want to use the backpropagation algorithm to update the weights and train the model. The question is, for each value of x1 and x2 there is a separate result (4 different combinations in total) and under different input pairs, separate error distances (the difference between the desired output and the actual result) could be be computed and subsequently a different set of weight updates will eventually be achieved. This means we would get 4 different sets of weight updates for each separate input pairs by using backpropagation.
How should we decide about the right weight updates?
Say we repeat the back propagation for a single input pair until we converge, but what if we would converge to a different set of weights if we choose another pair of inputs?
Now I understand that each neuron has certain weights. I am using a sigmoid function to determine a neuron should fire or not in each state.
You do not really "decide" this, typical MLP do not "fire", they output real values. There are neural networks which actually fire (like RBMs) but this is a completely different model.
This means we would get 4 different sets of weight updates for each input pairs by using back propagation.
This is actually a feature. Lets start from the beggining. You try to minimize some loss function on your whole training set (in your case - 4 samples), which is of form:
L(theta) = SUM_i l(f(x_i), y_i)
where l is some loss function, f(x_i) is your current prediction and y_i true value. You do this by gradient descent, thus you try to compute the gradient of L and go against it
grad L(theta) = grad SUM_i l(f(x_i), y_i) = SUM_i grad l(f(x_i), y_i)
what you now call "a single update" is grad l(f(x_i) y_i) for a single training pair (x_i, y_i). Usually you would not use this, but instead you would sum (or taken average) of updates across whole dataset, as this is your true gradient. Howver, in practise this might be computationaly not feasible (training set is usualy quite large), furthermore, it has been shown empirically that more "noise" in training is usually better. Thus another learning technique emerged, called stochastic gradient descent, which, in short words, shows that under some light assumptions (like additive loss function etc.) you can actually do your "small updates" independently, and you will still converge to local minima! In other words - you can do your updates "point-wise" in random order and you will still learn. Will it be always the same solution? No. But this is also true for computing whole gradient - optimization of non-convex functions is nearly always non-deterministic (you find some local solution, not global one).
Most examples of neural networks for classification tasks I've seen use the a softmax layer as output activation function. Normally, the other hidden units use a sigmoid, tanh, or ReLu function as activation function. Using the softmax function here would - as far as I know - work out mathematically too.
What are the theoretical justifications for not using the softmax function as hidden layer activation functions?
Are there any publications about this, something to quote?
I haven't found any publications about why using softmax as an activation in a hidden layer is not the best idea (except Quora question which you probably have already read) but I will try to explain why it is not the best idea to use it in this case :
1. Variables independence : a lot of regularization and effort is put to keep your variables independent, uncorrelated and quite sparse. If you use softmax layer as a hidden layer - then you will keep all your nodes (hidden variables) linearly dependent which may result in many problems and poor generalization.
2. Training issues : try to imagine that to make your network working better you have to make a part of activations from your hidden layer a little bit lower. Then - automaticaly you are making rest of them to have mean activation on a higher level which might in fact increase the error and harm your training phase.
3. Mathematical issues : by creating constrains on activations of your model you decrease the expressive power of your model without any logical explaination. The strive for having all activations the same is not worth it in my opinion.
4. Batch normalization does it better : one may consider the fact that constant mean output from a network may be useful for training. But on the other hand a technique called Batch Normalization has been already proven to work better, whereas it was reported that setting softmax as activation function in hidden layer may decrease the accuracy and the speed of learning.
Actually, Softmax functions are already used deep within neural networks, in certain cases, when dealing with differentiable memory and with attention mechanisms!
Softmax layers can be used within neural networks such as in Neural Turing Machines (NTM) and an improvement of those which are Differentiable Neural Computer (DNC).
To summarize, those architectures are RNNs/LSTMs which have been modified to contain a differentiable (neural) memory matrix which is possible to write and access through time steps.
Quickly explained, the softmax function here enables a normalization of a fetch of the memory and other similar quirks for content-based addressing of the memory. About that, I really liked this article which illustrates the operations in an NTM and other recent RNN architectures with interactive figures.
Moreover, Softmax is used in attention mechanisms for, say, machine translation, such as in this paper. There, the Softmax enables a normalization of the places to where attention is distributed in order to "softly" retain the maximal place to pay attention to: that is, to also pay a little bit of attention to elsewhere in a soft manner. However, this could be considered like to be a mini-neural network that deals with attention, within the big one, as explained in the paper. Therefore, it could be debated whether or not Softmax is used only at the end of neural networks.
Hope it helps!
Edit - More recently, it's even possible to see Neural Machine Translation (NMT) models where only attention (with softmax) is used, without any RNN nor CNN: http://nlp.seas.harvard.edu/2018/04/03/attention.html
Use a softmax activation wherever you want to model a multinomial distribution. This may be (usually) an output layer y, but can also be an intermediate layer, say a multinomial latent variable z. As mentioned in this thread for outputs {o_i}, sum({o_i}) = 1 is a linear dependency, which is intentional at this layer. Additional layers may provide desired sparsity and/or feature independence downstream.
Page 198 of Deep Learning (Goodfellow, Bengio, Courville)
Any time we wish to represent a probability distribution over a discrete variable with n possible values, we may use the softmax function. This can be seen as a generalization of the sigmoid function which was used to represent a probability
distribution over a binary variable.
Softmax functions are most often used as the output of a classifier, to represent the probability distribution over n different classes. More rarely, softmax functions can be used inside the model itself, if we wish the model to choose between one of n different options for some internal variable.
Softmax function is used for the output layer only (at least in most cases) to ensure that the sum of the components of output vector is equal to 1 (for clarity see the formula of softmax cost function). This also implies what is the probability of occurrence of each component (class) of the output and hence sum of the probabilities(or output components) is equal to 1.
Softmax function is one of the most important output function used in deep learning within the neural networks (see Understanding Softmax in minute by Uniqtech). The Softmax function is apply where there are three or more classes of outcomes. The softmax formula takes the e raised to the exponent score of each value score and devide it by the sum of e raised the exponent scores values. For example, if I know the Logit scores of these four classes to be: [3.00, 2.0, 1.00, 0.10], in order to obtain the probabilities outputs, the softmax function can be apply as follows:
import numpy as np
def softmax(x):
z = np.exp(x - np.max(x))
return z / z.sum()
scores = [3.00, 2.0, 1.00, 0.10]
print(softmax(scores))
Output: probabilities (p) = 0.642 0.236 0.087 0.035
The sum of all probabilities (p) = 0.642 + 0.236 + 0.087 + 0.035 = 1.00. You can try to substitute any value you know in the above scores, and you will get a different values. The sum of all the values or probabilities will be equal to one. That’s makes sense, because the sum of all probability is equal to one, thereby turning Logit scores to probability scores, so that we can predict better. Finally, the softmax output, can help us to understand and interpret Multinomial Logit Model. If you like the thoughts, please leave your comments below.
I need help in figuring out a suitable activation function. Im training my neural network to detect a piano note. So in this case I can have only one output. Either the note is there (1) or the note is not present (0).
Say I introduce a threshold value of 0.5 and say that if the output is greater than 0.5 the desired note is present and if its less than 0.5 the note isn't present, what type of activation function can I use. I assume it should be hard limit, but I'm wondering if sigmoid can also be used.
To exploit their full power, neural networks require continuous, differentable activation functions. Thresholding is not a good choice for multilayer neural networks. Sigmoid is quite generic function, which can be applied in most of the cases. When you are doing a binary classification (0/1 values), the most common approach is to define one output neuron, and simply choose a class 1 iff its output is bigger than a threshold (typically 0.5).
EDIT
As you are working with quite simple data (two input dimensions and two output classes) it seems a best option to actually abandon neural networks and start with data visualization. 2d data can be simply plotted on the plane (with different colors for different classes). Once you do it, you can investigate how hard is it to separate one class from another. If data is located in the way, that you can simply put a line separating them - linear support vector machine would be much better choice (as it will guarantee one global optimum). If data seems really complex, and the decision boundary has to be some curve (or even set of curves) I would suggest going for RBF SVM, or at least regularized form of neural network (so its training is at least quite repeatable). If you decide on neural network - situation is quite similar - if data is simply to separate on the plane - you can use simple (linear/threshold) activation functions. If it is not linearly separable - use sigmoid or hyperbolic tangent which will ensure non linearity in the decision boundary.
UPDATE
Many things changed through last two years. In particular (as suggested in the comment, #Ulysee) there is a growing interest in functions differentable "almost everywhere" such as ReLU. These functions have valid derivative in most of its domain, so the probability that we will ever need to derivate in these point is zero. Consequently, we can still use classical methods and for sake of completness put a zero derivative if we need to compute ReLU'(0). There are also fully differentiable approximations of ReLU, such as softplus function
The wikipedia article has some useful "soft" continuous threshold functions - see Figure Gjl-t(x).svg.
en.wikipedia.org/wiki/Sigmoid_function.
Following Occam's Razor, the simpler model using one output node is a good starting point for binary classification, where one class label is mapped to the output node when activated, and the other class label for when the output node is not activated.