if I train a deep neural network with (say) 10 classes and I feed the network a completely different image, is it reasonable to expect that the output layer's cells will not activate much, so I will know there is no object of the trained classes in the image?
My intuition says "Yes", but is it so? And what would be the best approach to this?
Thanks
During the supervised training you usually assume that during training you get complete representation of the future types of objects. Typically - these are all labeled instances. In your case - there are also "noise" instances, thus there are basically two main approaches:
As multi-class neural network has K output neurons, each representing the probability of being a member of a particular class, you can simply condition on these distribution to say that the new object does not belong to any of them. One particular approach is to check if min(p(y|x))<T (where p(y|x) is the activation of output neuron) for some threshold T. You can either set this value by hand, or through analysis of you "noise" instances (with you do have some for training). Simply pass them through your network and compare what value of T gives you best recognition rate
Add another one-class classifier (anomaly detector) before your network - so you will end up with the sequence of two classifiers, first is able to recognize if it is a noise or element of any of your classes (notice, that this can be trained without access to noise samples, see one-class classification or anomaly detection techniques.
You could also add another output to the network to represent noise, but this probably will not work well, as you will force your network to generate consistent internal representation for both noise vs data and inter-class decisions.
The answer to your question very much depends on your network architecture and the parameters used to train it. If you are trying to protect against false positives, we can typically set an arbitrary threshold value on the relevant output nodes.
More generally, learning algorithms mostly take the form of “closed set” recognition, where all testing classes are known at training time. However, a more realistic scenario for vision applications is “open set” recognition, where an incomplete knowledge of the world is present at training time and unknown classes can be submitted during testing.
This is an on-going area of research - please see this Open Set Recognition web page for plenty of resources on the subject.
Related
I’m very new to machine learning.
I have a dataset with data given me by a f1 race. User is playing this game and is giving me this dataset.
With machine learning, I have to work with this data and when a user (I know they are 10) plays a game I have to recognize who’s playing.
The data consists of datagram packet occurred in 1/10 second freq, the packets contains the following Time, laptime, lapdistance, totaldistance, speed, car position, traction control, last lap time, fuel, gear,..
I’ve thought to use a kmeans used in a supervised way.
Which algorithm could be better?
The task must be a multiclass classification. The very first step in any machine learning activity is to define a score metric (https://machinelearningmastery.com/classification-accuracy-is-not-enough-more-performance-measures-you-can-use/). That allows you to compare models between themselves and decide which is better. Then build a base model with random forest or/and logistic regression as suggested in another answer - they perform well out-of-the-box. Then try to play with features and understand which of them are more informative. And don't forget about a visualizations - they give many hints for data wrangling, etc.
this is somewhat a broad question, so I'll try my best
kmeans is unsupervised algorithm meaning it will find the classes itself and it best used when you know there are multiple classes but you don't know what exactly they are... using it with labeled data just means you will compute the distance of new vector v to each vector in the dataset and pick the one (or ones using majority vote) which give the min distance , this is not considered as machine learning
in this case when you do have the labels, supervised approach will yield much better results
I suggest try random forest and logistic regression at first, those are the most basic and common algorithms and they give pretty good results
if you haven't achieve the desired accuracy you can use deep learning and build a neural network with input layer as big as your packet's values and output layer of the number of classes, in between you can use one or multiple hidden layers with various nodes, but this is advanced approach and you better pick up some experience in machine learning field before pursue it
Note: the data is a time series, meaning that every driver has it's own behaviour of driving a car, so data should be considered as bulks of points, with this you can apply pattern matching technics, also there are a several neural networks build exactly for this data (like RNN) but this is far far advanced and much more difficult to implement
There is a dataset of a plant that makes certain numeric outputs based on numeric inputs. The dataset contains the input values and output value for several years every 15 minutes.
Since it would be too expensive to model the physical properties of the system in software, I would like to create a model with Machine learning, which behaves as the system. When entering inputs, the model should provide output.
For the solution I have tested Feedforward neural network. The results are ok, but in some cases too inaccurate.
What other methods would be available for this problem?
If it's a time series task you could use the NARX architecture of a neural network or an LSTM network. Later is like the NARX a recurrent neural network. Matlab offers an implementation of the first one.
https://en.m.wikipedia.org/wiki/Nonlinear_autoregressive_exogenous_model
https://en.m.wikipedia.org/wiki/Long_short-term_memory
If you "simply" want to fit a polynomial to your data you could use basic linear regression with polynomials of different degree to see which one works best.
Note: It's not called linear because it's only able to fit linear models.
https://en.m.wikipedia.org/wiki/Linear_regression
Some other possibilities are kernel methods such as kernel ridge regression or SVR. Later one is based on support vector machines which usually perform quite well (at least for classification from my personal experience).
If you want to try SVR you can use a small but great lib called libSVM. Matlab also offers this.
The following link shows a comparison of this algorithms:
http://scikit-learn.org/stable/auto_examples/plot_kernel_ridge_regression.html
Edit: If i understand this correctly, it's a time series task if you want to predict the outputs of a future time t+1 from a given time t. Try the NARX model or the LSTM net.
I understand that Batch Normalisation helps in faster training by turning the activation towards unit Gaussian distribution and thus tackling vanishing gradients problem. Batch norm acts is applied differently at training(use mean/var from each batch) and test time (use finalized running mean/var from training phase).
Instance normalisation, on the other hand, acts as contrast normalisation as mentioned in this paper https://arxiv.org/abs/1607.08022 . The authors mention that the output stylised images should be not depend on the contrast of the input content image and hence Instance normalisation helps.
But then should we not also use instance normalisation for image classification where class label should not depend on the contrast of input image. I have not seen any paper using instance normalisation in-place of batch normalisation for classification. What is the reason for that? Also, can and should batch and instance normalisation be used together. I am eager to get an intuitive as well as theoretical understanding of when to use which normalisation.
Definition
Let's begin with the strict definition of both:
Batch normalization
Instance normalization
As you can notice, they are doing the same thing, except for the number of input tensors that are normalized jointly. Batch version normalizes all images across the batch and spatial locations (in the CNN case, in the ordinary case it's different); instance version normalizes each element of the batch independently, i.e., across spatial locations only.
In other words, where batch norm computes one mean and std dev (thus making the distribution of the whole layer Gaussian), instance norm computes T of them, making each individual image distribution look Gaussian, but not jointly.
A simple analogy: during data pre-processing step, it's possible to normalize the data on per-image basis or normalize the whole data set.
Credit: the formulas are from here.
Which normalization is better?
The answer depends on the network architecture, in particular on what is done after the normalization layer. Image classification networks usually stack the feature maps together and wire them to the FC layer, which share weights across the batch (the modern way is to use the CONV layer instead of FC, but the argument still applies).
This is where the distribution nuances start to matter: the same neuron is going to receive the input from all images. If the variance across the batch is high, the gradient from the small activations will be completely suppressed by the high activations, which is exactly the problem that batch norm tries to solve. That's why it's fairly possible that per-instance normalization won't improve network convergence at all.
On the other hand, batch normalization adds extra noise to the training, because the result for a particular instance depends on the neighbor instances. As it turns out, this kind of noise may be either good and bad for the network. This is well explained in the "Weight Normalization" paper by Tim Salimans at al, which name recurrent neural networks and reinforcement learning DQNs as noise-sensitive applications. I'm not entirely sure, but I think that the same noise-sensitivity was the main issue in stylization task, which instance norm tried to fight. It would be interesting to check if weight norm performs better for this particular task.
Can you combine batch and instance normalization?
Though it makes a valid neural network, there's no practical use for it. Batch normalization noise is either helping the learning process (in this case it's preferable) or hurting it (in this case it's better to omit it). In both cases, leaving the network with one type of normalization is likely to improve the performance.
Great question and already answered nicely. Just to add: I found this visualisation From Kaiming He's Group Norm paper helpful.
Source: link to article on Medium contrasting the Norms
I wanted to add more information to this question since there are some more recent works in this area. Your intuition
use instance normalisation for image classification where class label
should not depend on the contrast of input image
is partly correct. I would say that a pig in broad daylight is still a pig when the image is taken at night or at dawn. However, this does not mean using instance normalization across the network will give you better result. Here are some reasons:
Color distribution still play a role. It is more likely to be a apple than an orange if it has a lot of red.
At later layers, you can no longer imagine instance normalization acts as contrast normalization. Class specific details will emerge in deeper layers and normalizing them by instance will hurt the model's performance greatly.
IBN-Net uses both batch normalization and instance normalization in their model. They only put instance normalization in early layers and have achieved improvement in both accuracy and ability to generalize. They have open sourced code here.
IN provide visual and appearance in-variance and BN accelerate training and preserve discriminative feature.
IN is preferred in Shallow layer(starting layer of CNN) so remove appearance variation and BN is preferred in deep layers(last CNN layer) should be reduce in order to maintain discrimination.
I've been studying neural networks for a bit and recently learned about the dropout training algorithm. There are excellent papers out there to understand how it works, including the ones from the authors.
So I built a neural network with dropout training (it was fairly easy) but I'm a bit confused about how to perform model selection. From what I understand, looks like dropout is a method to be used when training the final model obtained through model selection.
As for the test part, papers always talk about using the complete network with halved weights, but they do not mention how to use it in the training/validation part (at least the ones I read).
I was thinking about using the network without dropout for the model selection part. Say that makes me find that the net performs well with N neurons. Then, for the final training (the one I use to train the network for the test part) I use 2N neurons with dropout probability p=0.5. That assures me to have exactly N neurons active on average, thus using the network at the right capacity most of the time.
Is this a correct approach?
By the way, I'm aware of the fact that dropout might not be the best choice with small datasets. The project I'm working on has academic purposes, so it's not really needed that I use the best model for the data, as long as I stick with machine learning good practices.
First of all, model selection and the training of a particular model are completely different issues. For model selection, you would usually need a data set that is completely independent of both training set used to build the model and test set used to estimate its performance. So if you're doing for example a cross-validation, you would need an inner cross-validation (to train the models and estimate the performance in general) and an outer cross-validation to do the model selection.
To see why, consider the following thought experiment (shamelessly stolen from this paper). You have a model that makes a completely random prediction. It has a number of parameters that you can set, but have no effect. If you're trying different parameter settings long enough, you'll eventually get a model that has a better performance than all the others simply because you're sampling from a random distribution. If you're using the same data for all of these models, this is the model you will choose. If you have a separate test set, it will quickly tell you that there is no real effect because the performance of this parameter setting that achieves good results during the model-building phase is not better on the separate set.
Now, back to neural networks with dropout. You didn't refer to any particular paper; I'm assuming that you mean Srivastava et. al. "Dropout: A Simple Way to Prevent Neural Networks from Overfitting". I'm not an expert on the subject, but the method to me seems to be similar to what's used in random forests or bagging to mitigate the flaws an individual learner may exhibit by applying it repeatedly in slightly different contexts. If I understood the method correctly, essentially what you end up with is an average over several possible models, very similar to random forests.
This is a way to make an individual model better, but not for model selection. The dropout is a way of adjusting the learned weights for a single neural network model.
To do model selection on this, you would need to train and test neural networks with different parameters and then evaluate those on completely different sets of data, as described in the paper I've referenced above.
I Wrote a simple recurrent neural network (7 neurons, each one is initially connected to all the neurons) and trained it using a genetic algorithm to learn "complicated", non-linear functions like 1/(1+x^2). As the training set, I used 20 values within the range [-5,5] (I tried to use more than 20 but the results were not changed dramatically).
The network can learn this range pretty well, and when given examples of other points within this range, it can predict the value of the function. However, it can not extrapolate correctly and predicting the values of the function outside the range [-5,5]. What are the reasons for that and what can I do to improve its extrapolation abilities?
Thanks!
Neural networks are not extrapolation methods (no matter - recurrent or not), this is completely out of their capabilities. They are used to fit a function on the provided data, they are completely free to build model outside the subspace populated with training points. So in non very strict sense one should think about them as an interpolation method.
To make things clear, neural network should be capable of generalizing the function inside subspace spanned by the training samples, but not outside of it
Neural network is trained only in the sense of consistency with training samples, while extrapolation is something completely different. Simple example from "H.Lohninger: Teach/Me Data Analysis, Springer-Verlag, Berlin-New York-Tokyo, 1999. ISBN 3-540-14743-8" shows how NN behave in this context
All of these networks are consistent with training data, but can do anything outside of this subspace.
You should rather reconsider your problem's formulation, and if it can be expressed as a regression or classification problem then you can use NN, otherwise you should think about some completely different approach.
The only thing, which can be done to somehow "correct" what is happening outside the training set is to:
add artificial training points in the desired subspace (but this simply grows the training set, and again - outside of this new set, network's behavious is "random")
add strong regularization, which will force network to create very simple model, but model's complexity will not guarantee any extrapolation strength, as two model's of exactly the same complexity can have for example completely different limits in -/+ infinity.
Combining above two steps can help building model which to some extent "extrapolates", but this, as stated before, is not a purpose of a neural network.
As far as I know this is only possible with networks which do have the echo property. See Echo State Networks on scholarpedia.org.
These networks are designed for arbitrary signal learning and are capable to remember their behavior.
You can also take a look at this tutorial.
The nature of your post(s) suggests that what you're referring to as "extrapolation" would be more accurately defined as "sequence recognition and reproduction." Training networks to recognize a data sequence with or without time-series (dt) is pretty much the purpose of Recurrent Neural Network (RNN).
The training function shown in your post has output limits governed by 0 and 1 (or -1, since x is effectively abs(x) in the context of that function). So, first things first, be certain your input layer can easily distinguish between negative and positive inputs (if it must).
Next, the number of neurons is not nearly as important as how they're layered and interconnected. How many of the 7 were used for the sequence inputs? What type of network was used and how was it configured? Network feedback will reveal the ratios, proportions, relationships, etc. and aid in the adjustment of network weight adjustments to match the sequence. Feedback can also take the form of a forward-feed depending on the type of network used to create the RNN.
Producing an 'observable' network for the exponential-decay function: 1/(1+x^2), should be a decent exercise to cut your teeth on RNNs. 'Observable', meaning the network is capable of producing results for any input value(s) even though its training data is (far) smaller than all possible inputs. I can only assume that this was your actual objective as opposed to "extrapolation."