7
The neural network is typically a set size once it's created. You'd have to create a network big enough for your data-set.
6
Non-Euclidian geometry can be generally boiled down to the phrase
"the shortest path between 2 points isn't necessarily a straight line".
Or, put in a way that lends itself very much to machine learning,
"things that are similar to each other are not necessarily close if one uses eucidian distance as a metric" (aka the triangle inequality doesn't hold)...
4
In many cases, a production-ready model has everything it needs to make predictions without retaining training data. For example: a linear model might only need the coefficients, a decision tree just needs rules/splits, and a neural network needs architecture and weights. The training data isn't required as all the information needed to make a prediction is ...
3
We can already observe information bubbles on social media, where the circle is that machine learns what content people like and give more similar content based on clicks and so on. From single wrong click you could enter a bubble and never come out if you don't take care or be aware.
This happens with humans, so same may apply to computers. Checking ...
3
I did a little search and couldn't find any database that has ground truth for aggressiveness. This means that you need to build yourself a database. This might be huge undertaking. Take thousands of messages, and classify them by hand whether they are aggressive or not. This part is quite labor intensive.
Second part is much easier at start but would be ...
3
There is not single answer to the vanishing gradient problem. However, there a few things that can help.
As mentioned in the comments, use of Rectified Linear Units (ReLU) as your activation function can help, since the it does not get saturated for large neuron inputs.
Next, careful choice of weight initialization can help avoid saturation, as well. See ...
3
I think the crucial point here is what you precisely mean by mislabelled. Google's image classifier will likely do a 'pretty good' job of retrieving images with the given subject included, but how strict or lenient your class requisites are is quite important. For example, if one of your classes is 'dog' there may be hundreds of images procured from scraping ...
3
Recommendation systems can be applied for anything, as long as you have sufficient training data.
The most important inputs to the recommendation system are not "audio files or video files".
Wikipedia gives you the list:
Similarity matrix
Item attributes
User activities and behaviours
User profile
https://en.wikipedia.org/wiki/Recommender_system
3
Speaking to the "collecting data" part of the question, I'll say this:
Keep in mind that not everything requires massive amounts of data. Consider also that large amounts of data about all sorts of things are available in the OpenData / LinkedData realms. Governments in particular are sources of massive amounts of data. See, for example, data.gov or Google ...
3
There are a ton of sample datasets our there you can play with. A bunch of good ones install with R in the datasets package. Luckily you can download them independently if you're not an R user. Try https://vincentarelbundock.github.io/Rdatasets/datasets.html
You might also be interested in the MNIST database which is one of the canonical databases used ...
3
There are a few.
Sloth
LabelBox
RectLabel
LabelMe
OpenCV also has some facilitation for annotating.
Annotation is the more common name used in software suites for tools that facilitate adding labels to images and frames of movies because. Simple categorization is of limited use in the real world, especially with moving pictures where action may be ...
3
I will give a more thorough answer.
There are 2 problems you might face.
1) Your neural net (in this case convolutional neural net) cannot physically accept images of different resolutions. This is usually the case if one has Fully-Connected layers, however if the network is Fully-Convolutional then it should be able to accept images of any dimension. ...
3
How can data augmentation reduce overfitting?
You write that you can already maybe see how data augmentation can help prevent overfitting in general, but it sounds a bit uncertain and it's still asked in the title of the question, so I'll address this first:
Generally, when we use Machine Learning for classification problems, we would ideally learn a ...
3
You are quite correct. If you have properly followed the Cross Validation procedure and selected the best model indeed, then you can use the CV set as the training set for the final model. And no it will not cause your hypothesis to worsen (for that set maybe, but not for new examples) if you have selected the model correctly. In-fact you may use the entire ...
3
So I am assuming that you are trying to detect a lego brick from the image. One idea is that you can use transfer learning. Leveraging a pre-trained machine learning model is called transfer learning. The underlying idea behind transfer learning is that one takes a well-trained model from one dataset or domain, and applies it to a new one. François Chollet ...
3
There are a few tools that you can use to annotate (or label) data. For example, labelme or Labelbox. Have a look at this question for more alternatives.
3
I do not know a specific tool that meets all the mentioned requirements. However, a long time ago, I had to do a very similar task of labeling tons of images into 10 classes. This is how I did this:
Used a very basic clustering tool to cluster images into clusters (I set the number of clusters larger than 10 as I new some classes have very different ...
2
A simple way to do it would be lexicograpical sentiment analysis. To do that, you'd need a list of words categorized with a score that reflects "friendly" vs "aggressive" sentiment. For an example of setting up a SA system using Spark, see this article. To do what you're talking about, substitute AFINN for a different dataset. You might have to create ...
2
The answer by Cem Kalyoncu mentions the difficulty of building a ground truth database for aggressiveness.
One alternative approach would be to attempt to operate at the concept level, which would allow the use of pre-existing ontologies such as ConceptNet.
Here's a paper that describes this technique.
2
A popular dataset is the fisher iris dataset. It consists of 150 samples each with a dimensionality of 4. You can find it at
http://archive.ics.uci.edu/ml/datasets/Iris
2
I would also look at Minimum Edit Distances such as the Levenshtein distance.
You could use a dynamic programming technique such as the Viterbi Algorithm.
If you don't have a dictionary to work against, you may want to train with a Markov Chain model using a known "good" text. The Viterbi Algorithm could be used again to solve the model for the text being ...
2
I'd personally be more inclined to try longstanding deterministic methods such as Damerau (for typing errors) or Soundex (for homonyms arising from transcribed speech). At the very least, I'd use those as a baseline for any more 'AI-based' approach.
2
The GA will require a fitness function, which means you need labeled data for comparison.
That conclusion is wrong. Yes, sometimes your fitness function will use labeled data. For example, if you want to train an XOR gate or any other known function.
However, there is arguably no advantage of training a function with neuroevolution versus backpropagation, ...
2
For most of the current use cases, where NNs are used in conjunction with images, the image quality (resolution, color depth) can be low.
Consider image classification for example. The CNN extracts features from the image to tell different types of objects apart. Those features are pretty independent from the quality of the image (in reasonable bounds). ...
2
StackOverflow answers with code snippets. This data needs some processing, because the description can be in the question (along with other notes) and along with the answer. But this dataset is very big.
Also take a look at CodeReview questions.
2
Eh, "I won't be using camera calibration." .. Not sure, what you mean.
1st
At first, imagine a sheet of paper laying there on the floor. And first to try to transform the paper sheet (i.e. with some text on it), the picture of the sheet angled by paralaxe to the straight view: Just a transformation by a matrix. It would be highly valuable to have there ...
2
First of all, you mention that you have categorical data. I don't see how you can define similarity so that you can also define the distance between the predicted value and the ground truth (error). You can do that only if the data are ordinal.
If you want to just classify between normal and anomalous points (binary classification), without caring about ...
2
You can cluster all your features in one matrix X, in which each line would be one element of the data set you want to construct, and each column would be a different feature of this element.
You construct then a Y vector containing the different target classes, where the i-th element will be the target class of the i-th X element.
For the following I ...
2
Yes, it is possible, and yes, it probably has been done before. Odds are, however, the person(s) who tried were disappointed with the results and forgot to tell others.
The reasons they might be disappointed could be any of the following:
Took to long to train
Even when fully trained, (or appeared to be), it did not give the expected output
Too sensitive ...
2
Yes! Unsupervised machine learning has absolutely been applied to youtube videos... To recognize cats!
Here's an article about it in wired. One of the leading ML researchers was Andrew Ng.
Only top voted, non community-wiki answers of a minimum length are eligible
