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

learn more… | top users | synonyms

1
vote
1answer
65 views

How do you validate your machine learning models?

I am wondering what approaches are commonly used for validating a classification or prediction models: Approaches that am using at the moment: Using truth-sets: - ROCs, Bootstrapping, Accuracy, ...
0
votes
0answers
15 views

Forensics in wireless networks, anomaly detection and beyond?

first i'de like to apologize if this is not the right place. Next year i'm gonna be working on my final project in computer security, i have to build a wireless forensics tool that can analyse a data ...
1
vote
1answer
23 views

Feature Selection - Mutual Information with response variable that takes three values

I am trying to calculate Mutual Information scores for Feature Selection. I have successfully implemented the Mutual Information to test each feature against the binary response variable. Each ...
0
votes
1answer
30 views

Interpretation C value in linear SVM

My C value is very low (close to 0). Does this mean that my feature (dimensions) have no real separative (and thus predictive) value? (As the SVM basically chooses to ignore the training data ...
1
vote
1answer
53 views

Simple SVM Question

For a linear SVM, the documentation tells me the formula is: $$ \frac{1}{2}w^Tw+C\sum\limits_{i=1}^l\xi_i$$ Please explain to me in layman's terms what w (and ξ) represent. Is w the distance to the ...
2
votes
1answer
47 views

large variables and low sample (p > n) problem: ridge , LASSO, PLS, PCR which is most suitable for predictions

I am trying see whether to go for ridge regression, LASSO or principal component regression (PCR) or Partial Least Squares (PLS) in a situation where there are large number of variables / features (p) ...
1
vote
0answers
14 views

Should the 'TPR of adaboost' be better than base classifiers'?

There are two different base classifiers which produce true positive rate (TPR) values 99.46% and 91.79%. When I use these base classifiers in adaboost, what should the new TPR be? Better than two of ...
2
votes
1answer
28 views

Multi-armed bandit in face of full reward information

I am new to this area of machine learning. I am just walking myself through UCB1 algorithm which seems to assume that the payoff can be learnt only for action that ...
1
vote
1answer
45 views

How does cross-validation and the Bayesian method overcome the overfitting problem?

I was told that cross-validation and the Bayesian method can overcome the overfitting problem. I was told that comparison of models in a Bayesian way is actually doing cross-validation... What is the ...
0
votes
0answers
19 views

Why do we need nested cross-validation for parameter selection?

A normal cross-validation finds the best parameters such that: for each parameters: cross_validate(algorithm(parameters)) model = Use Best Model Why do ...
0
votes
0answers
14 views

Feature learning with a deep learning aproach?

How to create a feature vector from text with a deep learning aproach?. Im new at this topic, could anybody advice me where to start and how to aproach this task?.
11
votes
2answers
137 views

Bayesian lasso vs ordinary lasso

Different implementation software are available for lasso. I know a lot discussed about bayesian approach vs frequentist approach in different forums. My question is very specific to lasso - What are ...
0
votes
1answer
25 views

How to : a brief intro to scaling and rescaling data ( inputs) for supervised learning algorithms

I understand the concept of scaling and that it improves results in SVM's and NN's. however I would like to find somewhere where is is explained, in easy "layman's terms" terms. of how it is done. I ...
1
vote
0answers
33 views

What are appropriate validation methods for a Bayesian network model with low sample size?

I am currently using a Bayesian network model with 20 variables and 210 data points, with 15 locations measured at 14 different time points each. There are also some restrictions on what types of ...
3
votes
2answers
33 views

What model would be appropriate for predicting electrical consumption given multiple (mostly) independent variables?

I have about 1000 samples worth of daily electrical consumption for a building. I'd like to build a predictor based on a number of observable inputs, including: daily temperature (continuous) hours ...
0
votes
0answers
18 views

Softmax regression or $K$ binary logistic regression

For a multi-class classification problem, we can use $K$ binary logistic classifiers, or one softmax regression classifier, so how to make the choice between the two? IMHO, the $K$ binary logistic ...
0
votes
0answers
7 views

Use a different loss function for cross validation in liblinear

I am trying to learn a L2 regularized Logistic regression model in liblinear. I need a way to specify the C parameter which I do by cross validation. However, the loss/accuracy measure in cross ...
0
votes
1answer
38 views

Approach for mapping consumer preferences

I have this web application where I need to map consumer preferences based on some input information and individual choices. My goal is to create a list of product recommendations and evaluate the ...
0
votes
0answers
7 views

Specifying the validation dataset for liblinear

I am trying to use the liblinear logistic regression model with L2 regularization. I don't want the training data to be splitted for the cross validation. I want to specify my own validation set for ...
0
votes
0answers
26 views

In Gaussian Processes, how to understand the hyper-parameters optimization?

I know that in a GP, hyper-parameters are optimized by maximizing the marginal likelihood. Could anyone explain this method to me please? Thanks for your help in advance.
2
votes
1answer
52 views

How to understand the log marginal likelihood of a Gaussian Process?

I'm trying to understand Gaussian Processes. Could anyone tell me: Why we need to use the log marginal likelihood? Why using log, the marginal likelihood can be decomposed to 3 terms (including a ...
0
votes
0answers
29 views

Interpretation of Linear SVM Coefficients [duplicate]

I’m building a model using Linear SVM from the Scikit-learn package in Python. I have found that Linear SVM performs much better on my training set than Logistic Regression. My question is, is there ...
1
vote
1answer
36 views

Meaning of latent features?

I'm trying to understand matrix factorization models for recommender systems and I always read 'latent features', but what does that mean? I know what a feature means for a training dataset but I'm ...
0
votes
4answers
105 views

Analysing a CV with machine learning

What would you use to categorise or rank a Resume/CV with machine learning? I would like to make particular CVs stand out to recruiters.
0
votes
0answers
14 views

A framework for comparing the performance of Expectation Maximization

I have my own implementation of the Expectation Maximization (EM) algorithm based on the following paper http://pdf.aminer.org/000/221/588/fuzzy_k_means_clustering_with_crisp_regions.pdf I would like ...
-1
votes
0answers
11 views

Naive Bays classifier showing results for precision, recall and F value is always 1

I am implementing Naive Bays text classifier using Weka. I have trained it with very few words (about 20). I am getting the result that precision, recall and f value all as 1. Is this possible? ...
0
votes
1answer
64 views

What data mining/machine learning approach to use for a scoring model?

Suppose I have a large data set with lots of features(attributes). And I'm tasked to build some kind of scoring model to rank certain objects with all these features. How do I go about doing this? ...
0
votes
0answers
21 views

Method to select meaningful features for nearest neighbor classification

i try to perform some k nearest neighbor classification in R. That for i want to select the most meaningful features to deal with the curse of dimensionality. I have already decided to use Mahalanobis ...
3
votes
3answers
53 views

Recognition of simple patterns and prediction

I have been doing supervised learning and classification with multilayer perceptron for some time. But now I need to use unsupervised learning to infer the presence of a pattern and I would need some ...
1
vote
0answers
29 views

Squared Error vs Absolute Error loss functions [duplicate]

The two most popular types of loss functions are 1) squared error: $L(y,f(x))=(y-f(x))^2$ --> best estimate is the $E(Y|x) $ 2) absolute error: $L(y,f(x))=|y-f(x)|$ --> best estimate is the ...
0
votes
1answer
50 views

How to compare models from different but related datasets?

I'm building regression models on four the different but related data set and at the end, I want to test the significance of models. Since my models are built in a different data set, it's not ...
2
votes
0answers
53 views

Mixture of probits: understanding truncated-based likelihoods

I am trying to implement a mixture model of probits to infer the best decision boundary for every latent subpopulation. When doing Gibbs sampling, we eventually have to compute $P(y^* | w_c)$ where ...
1
vote
1answer
18 views

Regression Gaussian Estimates instead of points

I am using a support vector regression is order to get estimates of a variable y. I want to receive a probability distribution of my estimates and not just point estimates. I want to predict Gaussian ...
3
votes
1answer
39 views

When is it appropriate to use PCA as a preprocessing step?

I understand that PCA is used for dimensionality reduction to be able to plot datasets in 2D or 3D. But I have also seen people applying PCA as a preprocessing step in classification scenarios where ...
0
votes
0answers
14 views

After applying SVD, how do I find which features from my original dataset were most significant?

I'm using MathNet.Numerics library in C# to find the SVD but the Sigma matrix gives no indication of which values correspond to which features. It simply lists them in the most significant order. ...
0
votes
0answers
27 views

Calculating the variance of a model?

I often hear about the bias-variance tradeoff to evaluate classifiers. Now I want to calculate them. I often compute the AUC of a binary classifier to evaluate its performance and do a 10-fold ...
2
votes
0answers
37 views

Some doubt in reading Machine Learning A Probabilistic Perspective ( chapter 3.2 )

When I am reading Murphy's Machine Learning A Probabilistic Perspective. In chapter 3.2. I have some doubt. I think the author want to express is two things. First, we can use Bayes formula to ...
2
votes
1answer
76 views

What is the difference between a multi-label and a multi-class classification?

What is the difference between multi-label classification and multiclass classfication. Speficially, what is the difference between a label and a class? Please provide a clear example. "Multiclass ...
0
votes
0answers
25 views

Usefulness of Z-normalization in Machine Learning

Z-normalization means rescaling the feature $X$ by subtracting the average $\mu$ and dividing by its standard deviation $\sigma$, i.e., $(X-\mu)/\sigma$. What is the usefulness of normalizing data ...
1
vote
0answers
23 views

Binary classification of dated text documents with seasonality

I have a collection of training documents with publication dates, where each document is labeled as belonging (or not) to some topic T. I want to train a model that will predict for a new document ...
7
votes
3answers
324 views

How is cross validation different from data snooping?

I just finished "An Introduction to Statistical Learning". I wondered whether using cross-validation to find the best tuning parameters for various machine learning techniques is different from data ...
1
vote
1answer
68 views

Analysis of Customer satisfaction surveys

I have customer feedback data about 2-3 products from 100 customers. Number of questions are around 160. I have data in excel format. Header row contains the question and row below contains the ...
2
votes
0answers
53 views

Rademacher bounds for unbounded loss functions

All common treatment of PAC bounds based on Rademacher complexity assume a bounded loss function (for a self-contained treatemnt, see this handout by Schapire. However, I could not find any result for ...
3
votes
0answers
61 views

Duda, Hart, Stork No Free Lunch Discussion

Please see this question regarding Duda, Hart, and Stork's No Free Lunch Thm Discussion Hi all, I was having trouble understanding the description of the NFL theorem in Duda, Hart, and Stork. My ...
0
votes
1answer
41 views

Predicting the impact point of a moving object

Suppose we have a moving object (a horizontal projectile motion as one of the most basic examples). Is there any way to predict where it will hit finally? Please note that I'm looking for a machine ...
0
votes
0answers
20 views

How much time is reasonable for training Restricted Boltzmann Machine?

I have implemented Binary RBM in Matlab. I am using 60000 images as an input to train RBM. It takes approximately 11.3 minutes. I used tic and toc functions to evaluate above mentioned elapsed time. ...
3
votes
1answer
66 views

Variance-covariance matrix for ridge regression with stochastic $\lambda$

In ridge regression with design matrix $X$, outcomes $y$, fixed regularization parameter $\lambda$, and errors $\epsilon\sim\mathcal{N}(0, \sigma^2I)$, the computations for the ridge regression ...
0
votes
0answers
13 views

GB-RBM unable to learn and generate simple 2D-Motions?

I am trying to apply a GB-RBM to a variation of the LASA Handwriting Dataset to be able to generate new examples. My dataset contains motion trajectories of simple shapes like 'S' shapes, spirals, 'C' ...
1
vote
0answers
22 views

Bayesian Perceptron - how can I generate many different perceptrons?

I am going to implement the Bayesian version of a perceptron that I read in the Statistical Mechanics of learning, by Engel-Van Den Broeck. The idea to improve the performance is to use many Gibbs ...
0
votes
0answers
32 views

Conditional or Joint Probability under Various distributions

In various statistical models the baseline equation (like in Naive Bayes $$\mathrm{classify}(f_1,\dots,f_n) = \underset{c}{\operatorname{argmax}} \ p(C=c) \displaystyle\prod_{i=1}^n p(F_i=f_i\vert ...