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.

Filter by
Sorted by
Tagged with
0
votes
0answers
7 views

Expected behavior of adversarial attacks on deep NN?

I am trying adversarial attack (AA) for a simple CNNs. Instead of the clean image, my simple CNN is trained with attacked images as suggested by some papers. As the training goes on, I am not sure if ...
0
votes
0answers
5 views

Survival analysis for predictive maintenance on one machine

I have date for failures on one engine, 10 failures in total over a several years period, does it makes sense to split the data at each failure event and use the data with 10 failures as if I have ten ...
0
votes
2answers
10 views

What model should be used for vector to scalar time series prediction

I'm new to machine learning and I'm trying to work on a problem and the data looks like the following: ...
3
votes
2answers
37 views

Why important features does not correlated with target variable?

I'm testing if there is a relation between top important features and correlation between those features and target. I'm working on the titanic dataset. I plot the feature importance (using ...
0
votes
0answers
3 views

How to Forecast Quarterly Dependent variable with Monthly Independent Variables with Random Forest

I want to forecast US GDP(quarterly variable) using 100+ monthly financial and macro variables with Random Forest. My issue is that I do not quite know how to interpolate the quarterly data into ...
0
votes
0answers
18 views

What use is a test set in a continuous training setting?

I have tuned and trained a model that predicts an aspect of customer behavior from behavior within the most recent 12 months. After initial training, generalization performance was estimated based on ...
8
votes
8answers
1k views

Why do parameters go untested in Machine Learning?

I finished up a machine learning (ML) course a while back. Everything was as an optimization problem. No matter what predictive challenge you face in ML, you're generally minimizing some objective (i....
1
vote
1answer
46 views

Does Wikipedia explain gradient boosting in wrong way?

Wikipedia's geral Gradient Boosting is: Friedman's Gradient Boosting is: Why wikipedia's gradient boosting fit h_m through pseudo-residuals while friedman uses line 4 to fit h? My question is not ...
0
votes
0answers
17 views

How to calculate sample size and prove non-inferiority/equivalence between two risk prediction models

I am in need of determining the appropriate sample size to allocate to a test set in order to prove equivalence/non-inferiority of two prediction models. My model is a deep learning based model (using ...
0
votes
0answers
5 views

VAE Error - Invalid reduction dimension 1 for input with 1 dimensions [closed]

I am trying to build a Variational AutoEncoder but I am getting the following error. ...
0
votes
0answers
9 views

GCN in different epochs

Graph Convolution network models each node as an aggregation of the target node and its direct neighbors. By stacking more GCN layers we can access to further neighbors. I want to know, does training ...
0
votes
0answers
27 views

how to make validation accuracy and validation loss more stable?

I have been working on a project lately and this project is about detecting dimples in SEM images of ductile fractures, e.g.: and what I mean by detecting dimples is detecting the border of each ...
0
votes
0answers
33 views

Need Accuracy and Loss Graphs Explanation

My training and validation accuracy graph are near to each other. The validation accuracy is like 0.05-0.3 more than training accuracy. And the training loss and validation loss are also near like ...
0
votes
0answers
28 views

Sampling in an n-dimensional volume defined by supporting experimental data

I have some experimental data with 5 input parameters that generates one output. There are a few hundred data points that cover a piece of the 5D volume. After training various machine learning models,...
0
votes
1answer
17 views

Variance of projected data

What's the expression for the variance of projection of $N$ data points onto a vector $\mathbf{v}$ which is not necessarily a unit vector? I could only find an expression when $\mathbf{v}$ is taken to ...
0
votes
1answer
25 views

What is the posterior distribution $p(\textbf{f} | \textbf{y})$ for a Gaussian Process regression?

What is the posterior distribution $p(\textbf{f} | \textbf{y})$ for a Gaussian Process regression? Suppose that $p(y_n |x_n, f) = N(f(x_n), \sigma^2)$ with prior on $\textbf{f} = [f(x_1), \ldots f(x_n)...
1
vote
0answers
16 views

Do larger numbers of hidden layers have a bigger effect on a classification model's accuarcy?

I trained different classification models using Keras with different numbers of hidden layers and the same number of neurons in each layer. What I found was the accuracy of the models decreased as the ...
0
votes
0answers
14 views

Given a dependent variable how can I determine the relative effect of several independent variables to its value?

Say I have a series of N observations, and for each observation, I have 4 variables: $x$, $y$, $z$, and $q$, where $q$'s value depends on at least two of the other ...
0
votes
0answers
45 views

How to derive EM algorithm for this regression model?

Given that $y = Xw + \epsilon$ where $y $ is $n\times 1 $, $w$ is $p\times 1$ vector and $X$ is $n \times p$ matrix of inputs $x_1, x_2, ... x_n$ each of which are $p\times 1$ vectors. Further $\...
0
votes
0answers
6 views

How can I improve accuracy in classifying 6 common interactions between two people in still images?

My project’s goal is to classify 6 common interactions between two people in images including Boxing, Facing, Handholding, Handshaking, Hugging, and Kissing. • Dataset (6 categories correspoding ...
1
vote
0answers
8 views

why does grid search cross validation give some value of mean square error for different values of c, gamma, epsilon in support vector machine?

I am using libsvm in matlab for time series prediction using support vector machine. when i use grid search cross validation to ...
0
votes
0answers
26 views

First Principal Component Direction

I am trying to derive the first principal component direction from the definition and need help in finding which step is going wrong. Here's my attempt: $\mathbf{X} \in \mathbb{R}^{N \times p}$ is the ...
0
votes
0answers
20 views

Is G-mean score appropriate for assessing imbalanced binary classification?

I read literature showing how metrics such Accuracy, ROC-AUC, F-1 score can be misleading for the evaluation of classifiers on imbalanced data. Is G-mean score suitable for this context?
-2
votes
0answers
15 views

Looking for unbiased scoring metrics for binary classification of imbalanced data [closed]

My original dataset is extremely imbalanced big data with just 0.01% positive cases. That's why I worked on small breast cancer dataset where I created a severely imbalanced to see the which metric is ...
2
votes
0answers
38 views

NN keeps averaging on my regression problem

i am trying to understand MC dropout by implementing variational dense layers such as in this link (except I am doing it on Matlab), and so I first try to verify that my model can regress without ...
1
vote
1answer
14 views

Can the k-nearest neighbor algorithm tell you how many clusters there are among predictors?

I recently did a short course on machine learning in R and found the k-means and k-nearest neighbor techniques extremely interesting. Forgive my naivete if this is all wrong, but it seems like the ...
1
vote
0answers
10 views

How to find best features using statistical tests?

I am new to Machine Learning and the first part where I am stuck is identifying which features are best for the model. I understand that usually correlation heat map shows the input and output ...
0
votes
1answer
16 views

Stopping criteria for episodic yet “continuous” optimization tasks?

The task I have consists of applying a selection of "rewrite rules" to some given input in some order in to minimize some property of the input. For example, let us say the input is a tree ...
0
votes
0answers
17 views

Maximizing Booking Completion Rate [closed]

I was given a list of time, latitude and longitude of pickup and destination of bookings, along with the status of the booking (Completed, Not Completed, No driver found). The metric that we want to ...
0
votes
0answers
14 views

Gradient Boosting vs Gradient Descent

The section 10.10.2 of ESL claims that the difference between gradient boosting and steepest descent is that The tree components $t_{m} = T(x_{1};\theta_{m}),...,T(x_{N};\theta_{m})$ are not ...
1
vote
0answers
13 views

LSTM Time Series Forecasting

I am working with Bitcoin block data and attempting to forecast target hashrate ~one month in advance. I have condensed the raw block-by-block data into daily data and calculating metrics like ...
0
votes
0answers
10 views

Can using a dependent variable as a feature ever reduce prediction accuracy [closed]

There is a possible correlation between some of my features at a deeper level but I cannot be sure of it. Is it better to leave all of them in or could that ever cause a reduction in accuracy? I am ...
0
votes
0answers
7 views

How does Keras Evaluate handle Multilabel Sigmoid Problems

As I am working on tuning a model multilabel densely connected model, I am realizing that I don't understand how the model is being evaluated. My current model has 20 potential labels and ends with a ...
0
votes
0answers
7 views

Why does XGBoost watchlist shows better metric scores than actual predictions?

So I have ran XGBoost in R some dozens of times now and always found it strange how when I set the "watchlist" parameter with test set, the metric scores it shows are always better than the ...
0
votes
0answers
6 views

Statistical test for comparing performance metrics of two regression models on a *single* validation dataset?

What is an appropriate way to apply a statistical hypothesis test for evaluating model performance metrics (e.g. MAE, MAPE, MSE) between two regression models based on a single holdout (validation) ...
0
votes
1answer
30 views

If extrapolation is bad then how is forecasting methods statistically relevant

There are lot of articles out there that talks about why extrapolation is a bad thing to do. My question is if the above is true , how are forecasting methods like forecasting the trend based on some ...
0
votes
0answers
3 views

How can I find the optimal time to update data mart?

I am assigned to find out the optimal or some reasonable time when to update a table in the data mart I am working on. The data that I need to update in the data mart are collected from air ...
1
vote
0answers
14 views

Why Negative Log Likelihood (NLL) is a measure of model's calibaration?

In the context of (multiclass) classification, I've read papers which imply that NLL is minimized iff the model is well-calibrated (outputing true probability for each class and not just confidence), ...
0
votes
0answers
11 views

Test accuracy > Validation accuracy and Validation accuracy > Training Accuracy for some epochs

I am doing some baseline model testing for Stock price direction prediction. Currently trying a deep MLP with Early Stopping. Initially I had divided the data as 60-20-20 and I was using the ...
1
vote
0answers
16 views

Can KNN be overfit by adding too many variables?

GLMs can be overfit by adding too many variables. However, since with knn you are not estimating parameters, surely you are not using up degrees of freedom. So would this mean that you cannot overfit ...
1
vote
0answers
31 views

duplicate feature with same data in logistic regression

this is not a duplicated question What happens if I train a model on a data set that includes a duplicated feature? we have some feature such as x(1) and the label is y(1). we add a new feature that ...
0
votes
0answers
20 views

Machine learning methods to learn from misclassified samples

I trained a deep learning model in classifying images and deployed it in an application which users get image's coarse label from the model and then they correct the label. I want somehow the model ...
1
vote
0answers
22 views

low sample size - causal study using multiple outcome analysis

I am trying to do a healthcare analysis project (observational research) using retrospective data for the first time. Let's say I would like to study the effect of drug A. The primary outcome is ...
1
vote
0answers
12 views

Problem with a dataset not being properly labelled

I have a labelled dataset but these classes are not perfect. Some classes should be combined into one, whilst others have too few data-points for training. My main concern is the former not the latter....
1
vote
1answer
24 views

What does a “flat region” of precision recall curve imply?

I am evaluating ML models (GBDTs) on various test sets using Precision-recall curve, and my goal is: within some precision range, get as high recall as possible. The precision-recall curves on most of ...
4
votes
1answer
47 views

Why Gradient Boosting is needed at all?

I am trying to learn from Hastie on boosting methods. A question has bothered me for weeks. The book describes Forward-Stagewise Fitting, that is, fit a weak learner to explain the residuals from the ...
0
votes
1answer
16 views

How to prepare a 2x2 Confusion matrix for binary classifier [duplicate]

Problem statement: Evaluate a binary classifier. There are 50 positive outcomes in the test data, and 100 observations. Using a 50% threshold, the classifier predicts 40 positive outcomes, of which 10 ...
1
vote
0answers
17 views

VC-Dimension of Axis Aligned Right-Angle Triangles and 5-points Convex Hull

I am having trouble proving the following fact about VC dimension of triangles. Consider right-angle triangles in the plane, with the the right-angle in the lower left corner. The hypothesis in our ...
0
votes
0answers
7 views

Can someone please explain the decoder attention step in the transformer framework?

I just wanted to clarify how the highlighted attention step in the decoder works. From looking at the original paper (Vaswani, 2017 - "Attention is all you need"), I gather that the encoder ...
0
votes
0answers
25 views

Difference between Self-Taught learning and Transfer Learning?

I am new to Transfer Learning and I start by reading A Survey on Transfer Learning, and it stated the following: according to different situations of labeled and unlabeled data in the source domain, ...

1
2 3 4 5
337