Methods and principles of building "computer systems that automatically improve with experience."

learn more… | top users | synonyms (1)

3
votes
0answers
42 views

Bayesian interpretation of the repeated sampling principle

My question is philosophical rather than practical, and I will try to explain it through an example: Consider a Kaggle competition. All these contests have a similar structure: A "train dataset" is ...
1
vote
0answers
36 views

What is the correct equation of AdaGard one should use if one aims to use AdaGrad in practice as the automatic way to choose the step size?

I was reading Duchi et al. AdaGrad paper and also a shorter paper on Adaptive Online Gradient Descent because I was looking to implement an update rule which was automatically chosen and that was ...
1
vote
0answers
20 views

Greedy Training of Deep Belief Networks

I try to understand the justification of Greedy Training for Deep Belief Networks. I read the tutorial at http://deeplearning.net/tutorial/DBN.html and various papers of Hinton,Bengio and other ...
1
vote
0answers
27 views

numerical attributes with random forest

I am working on a problem of COPD exacerbation likelihood prediction, I have a total 62 attributes out of which 38 variables/attributes are of numeric/continuous type and remaining are either binary ...
0
votes
0answers
14 views

Is the following a standard HMM variant?

I have a problem that looks to me like a HMM variant. Could somebody confirm that I am on the right track modelling this and possibly tell me the name of this HMM variant if it is a standard HMM ...
3
votes
2answers
104 views

Diagonal in ROC plot?

I've used the R ROCR package to generate an ROC plot. I can see a diagonal in my plot. However, the ROC curve that I see other people generated has only horizontal and vertical lines. Why would I get ...
0
votes
0answers
23 views

Automatic Feature Extraction in Caffe

I am a beginner to frameworks like Caffe, Theano for deep learning. But, I would like to know how do I use Caffe framework for only feature extraction without performing classification. I want the ...
1
vote
0answers
36 views

Statistical testing in machine learning [closed]

I have a clarification. I am currently working as a data scientist for an IT company. I have experience in the area of NLP, ML and Big data for more than 5 years. I am from Bioinformatics background. ...
0
votes
2answers
24 views

Cluster two feature samples with no knowledge of the number of clusters

Thanks in advance for the help I have around 13000 samples with two features each and I would like to cluster these samples into groups. A few caveats. One, I don't know how many groups there are ...
0
votes
1answer
14 views

Using SVM's to output binary 0 or 1 to data

I am using a very large ~700,000 sample training set and ~700,000 sample testing set and training an SVM with the training set. When I run the SVM (SciKit-Learn) on the testing set it outputs only ...
1
vote
1answer
25 views

Machine learning analysis of differences

I'm very new to machine learning, and have a problem I'm trying to solve but don't know where to get started. I've got a set of feature vectors that are "matched pairs," so ...
0
votes
0answers
15 views

measure the convergence rate of SVM under cross-validation

I want to measure the convergence rate of my SVM based classifier, i,e how many data points are necessary to build an accurate classifier. I have an initial learning set from which I remove X% (X from ...
0
votes
0answers
14 views

Calculate personal server rating based on statistics and experience

There is a system with number of VPN-servers in different countries and users (which are also in different countries) which connect to that servers using number of available protocols. We gather ...
1
vote
1answer
58 views

Questions about understanding convolutional neural network (with Tensorflow's example)

I've recently seen tutorials from google's tensor flow. It was about convolutional network and way of convolution way slightly different from what I first learned. ...
0
votes
1answer
8 views

Neural Network border condition or extrapolation data

I was reading the user guide of NN for Matlab and I found this quote about extrapolation data: It is important that the data cover the range of inputs for which the network will be used. ...
0
votes
0answers
35 views

Use AUC for model comparison but what is the optimal threshold for final prediction

We can compare the performance of different models using AUC ROC and pick the one with large AUC. Then, we still need to choose and use specific threshold to predict the label for the test data. I ...
-1
votes
1answer
32 views

Independent and conditionally independent

I was wondering if two variables can be independent and conditionally independent. For example, A and D are independent. But are they also independent given the evidence C? I think they are, because ...
4
votes
2answers
84 views

In TensorFlow's Computational Model, is it possible to implement general machine learning algorithms?

https://www.tensorflow.org/ All the projects on TensorFlow I have seen in GitHub implement some kind of Neural Network model. Given TensorFlow is an improvement over the DAG (it's not acyclic any ...
2
votes
0answers
31 views

What systems are based on the following paper? [closed]

http://papers.nips.cc/paper/3150-map-reduce-for-machine-learning-on-multicore.pdf This paper has Andrew Ng as one of the authors. I really liked the idea presented in this paper, and when I looked ...
0
votes
0answers
15 views

Using cross validation to decrease variation in prediction outcomes

Im trying to predict the won category of some soccer matches. I have the following df: ...
0
votes
1answer
14 views

Evaluating a clustering against multilabels

I have a clustering of text documents, where each document is uniquely assigned to a cluster. I have a set of labels (keywords) attached to each document. That is, each label may be applied to many ...
2
votes
1answer
27 views

Is it a good idea to use $\eta = \arg \max( L(y, f(x) )$ to choose the step size for stochastic gradient descent (SGD)?

I wanted to have a good (as optimal as possible) automatic way of choosing the step size for minimizing the generalization error $\mathbb{E}_{ (x,y) \sim p_{x,y} }[L(y, f(x))]$, where $L$ is the loss ...
0
votes
0answers
28 views

Find sample complexity given dataset (hypothesis space)

In preparation for upcoming exams I'm trying to solve some exam questions from previous years. However, I got stuck on this following one: Assume given a dataset with n predictive attributes $x_1 ...
2
votes
1answer
67 views

When to use RBF networks instead of multilayer percetron?

I understand that a radial basis function neural network (RBF) usually has 1 hidden layer, and it differs from a multi-layer perceptron (MLP) via its activation and combination functions among other ...
1
vote
1answer
58 views

Find covariance if given mean and variance

I have a signal x that I want to classify in one of the classes A and B in which the means are Ma=[0.5,0.6] and Mb=[2,2] and with variances ...
0
votes
1answer
39 views

Text Classification for (Short) Open-Ended Survey Responses?

I am new to text mining/classification, and really want to learn more from this community. My data are open-ended survey responses (n=about 6,000) in which the respondent described what happened at ...
0
votes
0answers
8 views

Score based training libraries

Suppose I have a machine learning problem. However, the evaluation of the function that we are trying to predict is quadratic weighted kappa and it cannot be evaluated on one single data point. It is ...
0
votes
2answers
48 views

How to create a variable that is present in test data set but not in train?

Im try to do a classification but i have a variable production budget which is present in test dataset and not in train. so how do i proceed. could i impute that variable somehow. i dont want to drop ...
0
votes
0answers
19 views

Probability distribution in RandomForest

I'm trying to build a model with RandomForest in R, but I'm facing some issues. I'am using: ...
0
votes
1answer
37 views

Dealing with noisy features in a classification problem

I'm working on a classification problem whose features are very noisy. I have a table with the 'official' feature levels, but the actual data loosely resemble them. For example, to represent a value ...
0
votes
0answers
14 views

want to evaluate the performance of my model

I have developed a machine learning model to predict outcome of cricket matches and the accuracy I am getting from my model on the test set is 65% and on the training set is 66%. I want to know if it ...
0
votes
0answers
19 views

Is fitting hyperparameters to data in a Machine Learning model appropriate?

I have constructed a machine learning model (it is similar to Naive Bayes) within the Bayesian framework, and as such, have must select priors. In my brief exposure to Bayesian statistics, I was ...
0
votes
0answers
33 views

How to detect outliers?

I have a matrix where the rows are the data points (samples) and the columns are the features (predictors). Let's say I have 1000 data points and 20 features, i.e. the matrix is of size 1000 x 20. ...
1
vote
0answers
28 views

How to combine perceptron weights and probabilities

I am working on machine transliteration using structured perceptron to learn the model parameters at word level. I am using beam-search to extract the top $n$ transliterations of a word and would like ...
1
vote
1answer
36 views

Kernel nonparametric regression

One of the methods for nonparametric regression is using kernels. My question is what are the conditions on the kernels functions in this method? In other words how can I decide if a given function ...
0
votes
1answer
32 views

Linear Regret for epsilon-greedy algorithm in Multi-Armed Bandit problem

I am reading about $\epsilon$-greedy algorithm in Multi-Armed Bandit (or $K$ armed bandit) problem, as can be seen here: https://en.wikipedia.org/wiki/Multi-armed_bandit#Semi-uniform_strategies. For ...
1
vote
0answers
28 views

Difference in execution times between caret and randomForest (even with method = “none”)

I have a dataset with 1205 observations and 285 predictors (all but one categorical). It is a binary classification task. When I run randomForest, it executes in less than 1 second. When I run ...
0
votes
0answers
41 views

Curve Fit with logarithmic Regression in Python

I need to find a model which best fits my data. It's look like this: So I thought about logarithmic regression. But when I try to make a simple fit in python I get the following result: My code ...
1
vote
1answer
20 views

On skewed distributions, should the test/train/cross validation set be similar to the real set

On a skewed distribution (say there are two classes and the distribution is 2%-98%), given the small number of examples of one of the classes, would it be correct to have them distributed 50%-50% for ...
0
votes
0answers
35 views

Matching algorithm using probabilities and counts

I have a data set of id’s with location and time information. I want to determine which location is the home address of the id. For example, Id 1 could have the following data where percentage ...
0
votes
0answers
29 views

How to understand the result of label propagation algorithm

The graph is an undirected graph, “Protein 4","Protein 5", "Protein 6",... represents the vertex of graph. The value of two vertices with connected is 1 and that without connected is 0 in the ...
0
votes
1answer
53 views

Is that correct about dimensionality reduction and clustering?

Could you please help me to understand it because I'm not sure if I got it correctly. Let's say I have a dataset, of persons, with 100 features, various characteristics like height, weight, age, etc. ...
0
votes
0answers
13 views

Inference in Bayesian network Using bnlearn package

In this link Prediction of continuous variable using "bnlearn" package in R , the author talk about how I can find the conditionl probability of P(node(C)\ the rest node)=P(C\A,B,D,E,F,G) ...
1
vote
0answers
19 views

What is the “expressive power” of the composition function in a Recursive Neural Tensor Network?

I wasn't entire sure if this question belongs here or on math.stackexchange, since it's only partially about machine learning, but perhaps more of a conceptual question about linear algebra (or ...
0
votes
0answers
18 views

Continuous HMM library/Tool for continuous gesture recognition

There are many HMM tools, but most of them only support discrete density HMM (like Matlab). even the tools that support continuous density HMM can only be used for discontinuous gesture recognition. ...
1
vote
1answer
32 views

How the Markov Chain Monte Carlo ensure the stationary distribution converge to the target distribution?

I am reading about the MCMC but now I got a lot of questions. Firstly, It says we could construct a markov chain which satisfy the detailed balance:...
0
votes
0answers
126 views

XG Boost vs Random Forest for Time Series Regression Forecasting

I am using R's implementation of XGboost and Random forest to generate 1-day ahead forecasts for revenue. I have about 200 rows and 50 predictors. (As I go further in time I have more data so more ...
1
vote
1answer
50 views

Accuracy of training sample in Random Forest model in R

I'm using a Random Forest algorithm in order to construct a classification model, and I HAVE to check the accuracy of my rf model in the training sample, but as you can see in this answers : ...
0
votes
2answers
55 views

What algorithm should I try to predict number of tweets

Background I have been collecting data of each tweet being sent out for each article. The data is like this. created_at, user_id, article_id xxx, x, x y, y, y ...
0
votes
1answer
44 views

How to train a neural network on large images?

I am working through ideas for a computer vision problem, and most of the examples I see online use extremely small images, often 32x32 pixels. Does it make sense to try to train a neural network on ...