Stack Exchange Network

Stack Exchange network consists of 175 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

Questions tagged [machine-learning]

Machine learning algorithms build a model of the training data. The term "machine learning" is vaguely defined; it includes what is also called statistical learning, reinforcement learning, unsupervised learning, etc. ALWAYS ADD A MORE SPECIFIC TAG.

0
votes
0answers
12 views

Creating multivariate regression model out of multiple univariate models

A bunch of ML regression models are defined only for predicting the value of a single variable. Or have standard implementation that are only for the univariate case. For example support vector ...
0
votes
0answers
4 views

Event prediction with quantitative Social Media Data (Time Series)

I want to use Twitter data to predict a boolean event. Let's say the Event is labeled as 1 (no event = 0). I have several twitter datasets containing at least one timeframe with the event 1. A ...
2
votes
0answers
10 views

Why do we use the average-pooling layers instead of max-pooling layers in the DenseNet?

In the paper of DenseNet. The author adopted average-pooling in the transition layers. So what is the motivation of such choice? Why not using max-pooling layers?
0
votes
0answers
12 views

How reduce categorical level and group labels by finding duplicates [on hold]

In a classification multi class problem with high level categorical data , what can be used to reduce the categorical level of features or reduce amount of labels? my data contains duplicates levels ...
0
votes
0answers
22 views

Getting sets of random correlated variables

For the training of a machine learning model I need to add additional features, and these features are correlated. I need to run the model N times adding these features with random values, and for ...
0
votes
0answers
10 views

which teams in the Nigeria npl lost their last matchs and will likely win their next match [on hold]

Which teams lost their last match in the Nigeria professional league and will likely win their next match
1
vote
0answers
6 views

Transfer Learning in domains other than Image processing and NLP

Can Transfer Learning be applied in domains other than Image processing or NLP? I am trying to apply it on clickstream data (for propensity modeling). Any reference would be greatly appreciated.
-1
votes
0answers
11 views

I need kmean++ algorithm [on hold]

Can someone tell me the exact algorithm for kmean++? What i have understood is the following: 1) choose an initial center c_1 uniformly at random from X. Compute the vector containing the square ...
1
vote
1answer
43 views

Missing data when actually predicting - Additional model legit?

This is a theoretical question, but I have already stumbled upon this issue a few times: My learning data are not complete, but I manage to handle the missing values. Now it's time for actually ...
0
votes
0answers
6 views

Delving Deep into rectifiers:surpassing human-level performance on ImageNet classificatiom

I have read this paper (Delving Deep into rectifiers:surpassing human-level performance on ImageNet classificatiom)again and again and again ,but I do not understand initialization of filter weights ...
0
votes
1answer
8 views

Can I use the Silhouette to measure quality of clusters in different dimensions?

Can I use the Silhouette to measure quality of clusters in different dimensions? For example, let's say we run kmeans for some $k$ using 6 features of the dataset. Mark the resulted silhouette as $...
-1
votes
1answer
31 views

Statistical technique to combine scores from multiple tests

In my experiment I have two groups, one group has an anxiety disorder (N = 22), the other does not (N = 11) (I know low sample size). They were tested in a battery of psychological tests and also ...
0
votes
0answers
17 views

Opposite results from the residuals and JB result test?

I`m trying to forecast some forex returns of currencies couples. I build up my ARIMA model and test for normality of distribution after the arima is applied. I get different results from the Jarque - ...
0
votes
0answers
9 views

Autoencoder loss function - why minimise MSE?

Why are most loss functions used in autoencoder learning algorithms the mean squared error?
0
votes
0answers
12 views

Random oversampling versus classes weighting for class-imbalanced dataset [on hold]

Between random oversampling and classes weighting, which is better to balance the dataset? Why?
1
vote
0answers
5 views

Variation in accuracy of data splitting before and after data augmentation

How much accuracy of the system varied/changes between two cases Data augmentation before splitting Data augmentation after splitting, only on training data Is there any literature published?
1
vote
0answers
8 views

Same input size for Source and Target Model?

For the transfer learning do we need to have same input image size of target model as source model? for example my source model is trained on 100x100 input images and target model is low resolution ...
0
votes
0answers
8 views

Kullback-Leibler divergence Loss With different-length vectors

I am new to KL Divergence Loss (and indeed all similar comparisons between discrete series data). The output of my network produces a series of tuples of a length that varies during training. The ...
0
votes
1answer
14 views

Loss function for ordered categorical data [on hold]

I am using pytorch and I am working on a test problem where I have 10 output categories, but they are ordered (in an image segmentation problem). That is, 9 is closer to 10 than it is to 1. I want to ...
1
vote
0answers
10 views

What do we call models that output probabilistic scores vs. models that output non-probabilistic scores?

In machine learning -- specifically in binary classification -- there are models that output a probability for each data-point fed to the model. For example, a logistic regression could take some ...
0
votes
0answers
14 views

How to compute weight change for hidden layers with cross-entropy loss? [duplicate]

I'm trying to train a neural net with 1 hidden layer (RELU) softmax output layer cross-entropy loss stochastic gradient descent My implementation seems to work fine when I don't use any hidden ...
0
votes
0answers
12 views

What's the effect of using TF-IDF encoded instead of one-hot encoded categorical data as input to a neural network?

As input into a simple neural network multi-class classifier, I am considering using a variation of the standard one-hot sparse matrix to represent categorical variables. Instead of each element ...
0
votes
1answer
21 views

Data augmentation on entire dataset before splitting

If I apply rotation of 5 different angles and randomly cropp 10 different images from each rotated image and than divided into training testing and validation. Will it be totally incorrect evaluation ...
0
votes
0answers
11 views

Can someone provide a brief explanation as to why reproducing kernel Hilbert space is so popular in machine learning?

I thought functional analysis was long thought to be old fashioned and generally a dead research area. It seems that all of a sudden there is a huge fascination with so-called reproducing kernel ...
3
votes
1answer
22 views

Best activation and loss function for regression problem where outputs are from 0 to 1

I'm currently working on a regression problem, where the targets are from 0 to 1. Which would be the best pair of activation and loss function for these kinds of problems? The ones that I have ...
0
votes
0answers
23 views

Is the model over-fitting the data?

On the y-axis you've got RMSE and on the x-axis you've got the number of epochs. Then in blue, the validation error, in red the training error. What do you think is the optimal number of epochs ...
0
votes
0answers
14 views

Could this data be generalized by machine learning? [on hold]

Please bear with my english.. Hi. I'm a graduate student in South Korea. These days, I try to build an appropriate model for data. The problem is classification algorithms(Gradient Boosting, ...
2
votes
1answer
41 views

Learn this distribution from samples? What is the sample complexity?

$\newcommand{\norm}[1]{\left\lVert#1\right\rVert}$ We have an $n$-variate distribution $X\in\{0,1\}^n | \sum_i^n X_i = k$. Or, in other words, we are guaranteed that only $k$ variables will be $1$ in ...
0
votes
1answer
18 views

Accuracy of SVM prediction

I'm trying to build a text classification model with SVM. The training data set consists of 100 string records with a one-to-one mapped response variable which is also a string. I can't split the data ...
0
votes
0answers
9 views

Question About Preprocessing And Regression Models

My first question is weather or not the head of data below needs to be further processed, or if the format is good the way it is. ...
1
vote
0answers
12 views

Is it redundant to normalize test data during Mahalanobis calculation?

I have a training set which is a 12-column dataframe that I'm using to generate a covariance matrix (and possibly a PCA model). My test set is a single 12-column vector. All 12 variables have the ...
0
votes
0answers
20 views

What's the bug in my implementation/understanding of backpropagation?

For learning purposes, I'm trying to implement a simple neural network with only linear layers followed by logistic activation. As far as I understand, the backpropagation algorithm exploits the ...
0
votes
0answers
6 views

Cross-Entropy Cost Function : Why to apply partial derivative twice? [migrated]

Here my question is Why do we need to apply partial derivative twice on the cross entropy cost function? Why not only single time ? Here is the image for reference:
0
votes
0answers
8 views

Calibrate conditional probability from one classifier to the conditional probability from another classifier?

I have two data sets with conditional probabilities predicted from two classifiers. Every data set has only two classes, say 0 and ...
4
votes
0answers
31 views

How to reconstruct an image from a training set?

Description: I have taken a series of images/photos of a panorama from different positions (x,y) in space pretty close to each other (max 100m difference). Here there is a top view representation to ...
0
votes
0answers
8 views

Is there a bias (atomistic fallacy) when I dis-aggregate higher-order variables to the individual level when using k-means clustering? [on hold]

I want to perform a k-means clustering, but I have a nested structure. However, I didn't find any packages supporting multilevel k-means clustering in R (or Python). Therefore, I'm thinking about dis-...
-2
votes
0answers
30 views

NEURAL NETWORK IN R [on hold]

MY ERROR IS CoMING TOO LOW WHILE I TRAINED THE DATA IN R USING NEURAL NETWORK BUT WHEN I AM TESTING IT IT IS GIVING NO CORRELATION BETWEEN PREDICTED AND ACTUAL DATA VALUES.Please help. i think there ...
1
vote
1answer
19 views

Calculating per cent improvement in test error

I’ve come across this statement regarding an improvement percentage and I don’t know if it’s okay. It says this: The training error and test error using the neighborhood model are 0.34 and 0.54 ...
0
votes
1answer
19 views

Practical Examples: Expectation of a function with respect to a probability

I have encountered the following phrasing while reading Bishop's "Pattern Recognition and Machine Learning": Although for some applications the posterior distribution over unobserved variables will ...
0
votes
0answers
10 views

In Xgboost, how does Scale_pos_Weight work for regression?

I have noticed I am getting better results if use scale_pos_weight. The training data is imbalanced. I have tried sampling but didnt get good result. I have two questions: 1. How does scale_pos_weight ...
1
vote
1answer
20 views

Theory on custom loss functions for GBDT and other ML

I'm looking for resources on the theory behind choosing a loss function for ML---I'm interested in GBDT but for deep learning would work as well. I'd like to get a better understanding of how the loss ...
0
votes
1answer
26 views

Detect Typical Customer Mistakes in the Shopping Cart [on hold]

I would like to ask your advice on solving this problem. Problem There is an online store which sells furniture. There are millions of products on the store like furniture parts and furniture itself....
0
votes
0answers
12 views

What are bias and variance for a resampling scheme (as opposed to bias and variance for a model)?

I am trying to understand how bias and variance of a resampling scheme like cross validation are different from bias and variance of a model (like a linear model or a tree). Which of the following is ...
0
votes
0answers
13 views

Efficient algorithm for finding optimal number of breakpoints for piecewise regression

I aim to implement a module in python that does the following: 1) Upon taking training data, fit a piecewise regression with $n$ breakpoints. 2) Determine how well of a fit it is to the data (I ...
1
vote
0answers
17 views

Multiple Non-linear Regression with Function-based regression and Machine Learning models

I'm working on an application of Multi-nonlinear regression. Initially, I tried this algorithmically by creating a polynomial of the form A(x^p * y^q * z^r). I saw ...
1
vote
1answer
44 views

Independent and dependent variables Machine Learning

I want to start a regression model on lottery numbers, my database consist of the dates, and results(numbers). In order to do a regression model I believe you need independent and dependent variables, ...
0
votes
0answers
10 views

Ensemble model for regression algorithms

I'M a Research scholar and working on regression models on my dataset, i have to apply ensemble model on my datasets for different regression algorithms, but i feel very difficult and i'am unable to ...
2
votes
1answer
44 views

is cost function of logistic regression convex or not? [duplicate]

For logistic regression, the loss function is convex or not? Andrew Ng of Coursera said it is convex but in NPTEL it is said is said it is non convex because there is no unique solution. (many ...
0
votes
0answers
10 views

Permuting RV order in stacked Auto-regressive Flows for density estimation

Brief background: Normalizing flows such as detailed in MAF and B-NAF use an auto-regressive formulation such that highly expressive bijective transformations of the RVs satisfy the probability chain ...
0
votes
0answers
16 views

Cross visitation using logistic regression [on hold]

I'm trying to understand this paper that cites using logistic regression to measure cross visitation. However, there isn't much information in the actual methodology. Could someone explain how to ...