Stack Exchange Network

Stack Exchange network consists of 174 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Visit Stack Exchange
Join us in building a kind, collaborative learning community via our updated Code of Conduct.

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

0
votes
0answers
15 views

Is Bias-Variance Trade Off and MSPE the same?

I've question regarding to the correct notation. I was reading about the Bias-Variance-Trade-Off in the textbook "Elements of statistical learning". Is the expected forecast error listed there the ...
0
votes
0answers
4 views

Right techniques to cluster/segment categorical data?

I am new to this forum and to data science. I might be naive in asking my question. I am working on customer transactions data. I have got data of ~143k customers; for each customer I have monetary ...
0
votes
0answers
9 views

Neural network regression: seemingly bounded output

I have been working on a neural network based predictor for a project. The aim is to learn a certain quantity, say the signal strength of a cellular network, for each coordinate set in the dataset. ...
0
votes
2answers
22 views

In CNN, do we have learn kernel values at every convolution layer?

I'm new to machine learning and one of the things I don't understand about CNN is whether we have to learn the kernel values at every convolutional layer, or just learn a single set of kernel values ...
0
votes
0answers
6 views

How do sample weights work in classification models?

What does it mean to provide weights to each sample in a classification algorithm? How does a classification algorithm (eg. Logistic regression, SVM) use weights to give more emphasis to certain ...
0
votes
0answers
7 views

Convert datetime in dataframe to seconds

Below is the output of [mydataframe].info() ...
0
votes
0answers
8 views

How to compare the training performance of a deep learning model on different data sets?

So I have a deep learning model and three data sets (images). My theory is that one of these data sets should function better when it comes to training a deep learning model (meaning that the model ...
1
vote
1answer
20 views

Conditional Expected Value

I'm trying to figure out the following case; suppose that I have a supermarket where I give points to my customers for they to redeem (1 point for 1 dollar). Once in a while I send offers to my ...
1
vote
1answer
39 views

Can neural networks learn $g(x)$ from $\mathbb{E}[g(X_t)] = \int_{-\infty}^{\infty} g(x)p_t(x)dx$

Let $\mathbb{E}_x[g(X_t)]$ be the expected value of a random variable $X_t$ with known probability density $f_t(x)$ then for the continuous case $$\mathbb{E}[g(X_t)] = \int_{-\infty}^{\infty} g(x)...
2
votes
1answer
16 views

Co-variate shift between train and test data-set

What are some of the modern techniques used to detect and mitigate the covariate shift between train and test datasets? One method I recently read involves training a classifier using Kullback-...
1
vote
0answers
11 views

Faster RCNN - Pyramid of Filters vs Pyramid of Anchors (Reference Boxes)

I'm reading faster RCNN paper now and trying to understand what is the difference between Pyramid of Filters and Pyramid of Anchors methods from the scale point of view. I mean if I use only one ...
0
votes
1answer
20 views

Gradient in batch-size

When we set a batch-size, after each sample of batch passed we take the gradient but wait until last sample of batch to passed and then propagate the sum of gradient of them through the network? Am I ...
0
votes
0answers
19 views

Which unsupervised machine learning algorithm should I use?

I am observing emails in our online shopping website email address. Some of the emails are spam because they contain hyperlinks, contain information which are not related to our products and some ...
0
votes
1answer
18 views

For multiclass classification purpose I have to use a imbalanced dataset

I am facing a problem. It's a multiclass classification problem I have 5 categories A has 107 instances B has 101 instances C has 882 instances, D has 229 instances and E has 129 instances. I used Knn,...
0
votes
1answer
29 views

Random Forests and Information gain

Suppose you are building random forest model, which split a node on the attribute, that has highest information gain. In the below image, select the attribute which has the highest information gain? ...
-1
votes
1answer
15 views

How to weight features when doing text mining?

I have a case where I'm doing text mining over a list of product titles. In particular I want to run a clustering algorithm. But I also have some information about those products that I think can add ...
2
votes
0answers
23 views

Similarity between Train and Test data sets

I have multiclass classification dataset and I am using Deep nets for the classification task. To explain the problem, let's assume that I have 5 classes to classify. No matter what I try, be it ...
0
votes
0answers
38 views

Algorithms for everyone [on hold]

Something I've allways wanted to see is a concise run-through of different machine learning algorithms, all on one page: With their pros and cons, what situations they work in best and when they don'...
1
vote
1answer
12 views

Why convoloution neural net have to find filter values ?

I'm new to ML stuff and one of the thing that I don't understand about CNN, is that why CNN have to find the values of filter at convolution layer, why don' they use existing filters and only find the ...
0
votes
0answers
19 views

Deploying an ML model [on hold]

I am developing a model for outlier detection. It is really basic and simple. It uses the 3-sigma rule to detect anomalies. But, I am having trouble productionizing it. Specifically, I want to know, ...
1
vote
0answers
19 views

What method should I use for comparing the performance of two classifiers statistically?

1) Based on this paper: Evaluating the Replicability of Significance Tests for Comparing Learning Algorithms by Remco R. Bouckaert and Eibe Frank, I conclude that "using 100 runs of random subsampling ...
0
votes
0answers
19 views

Feature selection in the presence of extremely large feature set

Suppose you have a very large feature set (1000s to 1000,000 features) when building a machine learning model. How do you go about selecting the features? I know of the following methods for feature ...
1
vote
1answer
18 views

Test set error estimation under a small sample size

What are some of the plausible approaches to estimate the test set error when you have a very small sample set available (specifically, when the small sample size does not justify using a hold-out set)...
0
votes
0answers
18 views

Divergence of regularized gradient descent [on hold]

I am applying a regularized gradient descent algorithm on a dataset for linear regression. Since there are too many features, I am programming using the matrix notations. Following expression is being ...
-1
votes
1answer
43 views

is it allowed to rerun your entire predictive model?

Hi i am new to r and I have made a predictive model for the iris dataset (using the multinom function from the package nnet). When i ran the model for the first time I got an accuracy score 0f 0....
0
votes
0answers
9 views

How do I make use of a Weka classifier on data extracted via the pcap library in java? [migrated]

So Far: Firstly I've been able to extract a large amount of data from packets either being read from a pcap file or live packets (via the pcap library) as they are transferred to or from my local ...
0
votes
1answer
20 views

Machine Learning model with aggregated data as training

I would like to predict a LABEL: A,B or C using a classification machine learning model. My data to train the model is like: ...
0
votes
0answers
18 views

Perplexity of a Non-Statistical Language Model

I have a piece of software that, given a input phrase, returns an ordered list of the next most likely words (entire vocab is ordered 1 to n). This is essentially an Language Model with the exception ...
1
vote
0answers
30 views

How to input a continuous distribution to a neural network

I have simulated the relative frequency of a stochastic process by creating a very small grid say $1000$ by $1000$. The graph looks like this Now I am trying to setup a regression model by ...
1
vote
1answer
25 views

Why ensemble of many deep-learning models did not work?

I am trying to solve an image classification problem using DL, Keras and tensorflow. I added several layers of conv2D followed by batchnorm, pooling and dropout. I get a good accuracy ~95% with this. ...
-1
votes
0answers
11 views

Methods to explain one variable by another variable?

I have two time series: cumulated stock market return and daily market sentiment scoring. I ran a simple linear regression (with or without intercept), but the R2 is quite low (0.37). Correlation ...
1
vote
1answer
29 views

Size of the Hypothesis Space

(I'm asking the same question as the linked one, I simply don't have enough reputation to comment yet, but hopefully, this one will more clearly explain what me and the other asker both mean) Let's ...
0
votes
0answers
17 views

(Re)-Train on a small dataset and new incoming data

I would like to train a classifier (doesn't matter which learning algorithm) on a small set of training data. As soon as the system predicts new samples, it should collect them, add the samples to the ...
2
votes
1answer
29 views

How does neural network training work, if there are A HUGE number of points that not differentiable?

When I first saw ReLu function, I would not guess it will work in neural network because there is a point that is not differentiable. But it seems works very well on modern neural network. My ...
1
vote
2answers
15 views

Is it better to avoid ReLu as activation function if input data has plenty of negative values?

ReLu is probably the most popular activation function in machine learning today. Yet, ReLu function outputs 0 when input data values are negative. ReLu totally disregards negative data. This may ...
1
vote
0answers
14 views

Probability calibration metric for multiclass classifier

A machine learning classifier can be calibrated so that when the probability that datapoint i is of class A is 0.6, this is true 60% of the time. In the binary class setting, this can be visualised ...
0
votes
0answers
16 views

extensions of logistic regression in the context of machine learning

I was wondering whether there exists an overview about all extensions of logisitic regression in the context of a machine learning approach. E.g. instance-based logistic rgression (Cheng and ...
1
vote
0answers
31 views

Equation for standard error of linear predictor and 95% prediction intervals from logistic regression

I would like to present 95% prediction intervals in an online risk calculator. 1) After fitting a logistic model with lrm (which includes some restricted cubic ...
0
votes
0answers
20 views

How to decide which activation function to use? ReLu or sigmoid? [duplicate]

Sigmoid function gives equal weightage to both negative and positive input values. ReLu function outputs 0 for negative input values and output follows the input for positive input values. Sigmoid ...
1
vote
0answers
15 views

GAN for learning the transition density of a Markov process

I have learned about the Generative Adverserial Networks and the way they are used for learning the underlying (complex) distributions of high dimensional data. Now, my question is: Are there ...
0
votes
0answers
12 views

How to add noise to supervised (binary-classifier)?

Note: The question is not about validating/testing a trained model. Say i have an unlabeled features set, I want to approximate the true labels (for the sake of argument lets assume it's a binary ...
9
votes
3answers
196 views
+150

What *is* an Artificial Neural Network?

As we delve into Neural Networks literature, we get to identify other methods with neuromorphic topologies ("Neural-Network"-like architectures). And I'm not talking about the Universal Approximation ...
0
votes
0answers
13 views

Binary Class Imbalance - not enough of class A or too much of class B?

Assume 1000 instances of both class A and class B are sufficient to train a decent binary classifier. Since it is easy to get more class B data, we get 99000 additional instances of class B (if we ...
3
votes
0answers
21 views

Is Gradiant Boosting a generalization of Adaboost?

I read somewhere that Gradiant boosting is a generalization of Adaboost. However, I cannot see why. Can Anyone elaborate?
0
votes
0answers
3 views

Model deployment - the right data batch structure

This question goes mainly out to the data scientist and data engineers that have applied experience, although anyones 2cents are appreciated. Intro: I have a production ready model that has been ...
0
votes
1answer
16 views

is there any diffrence to use nonlinear or liner activation function in single hidden layer network

I am working in a classification problem in which I use RBF with a single hidden layer. I want to use SoftMax activation function for the hidden layer. I already read some documents about the ...
-1
votes
0answers
8 views

Is it sufficient to normalize the input Data to perform a multiple Output Regression where the labels have different magnitudes?

I am trying to simplify a complex mathematical model in a certain range by performing a regression with a neural Network. I am using a hidden layer with a 'tanh' activation function to normalize my ...
1
vote
1answer
15 views

Adding the input layer - units with a decimal

I took the course Machine Learning A-Z from Udemy and am trying to apply what I learned in the tutorials. Theye taught us in the "Adding the input layer" portion of an ANN that the units is based off ...
0
votes
1answer
17 views

multiclass SVM classification (using R)

I'm new to supervised classification. Here's my case: I want to classify subjects in 3 classes: healthy, sick and intermediate. I've been asked to use SVM to do the classification. I know how it ...
0
votes
1answer
52 views

When was the first time that logistic regression was used to forecast an unknown outcome?

Logistic regression is originally used to predict probabilities of a binary response or further used to forecast the binary response for unknown responses based on a test data set. I was wondering ...