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.
19,990
questions
0
votes
0
answers
6
views
Search recall optimization - what appropriate loss function to use?
I am studying machine learning and wanted to work on a project of my own so that I have better chances after graduating college. I'm studying the application of ML to improve searches using a toy ...
- 161
1
vote
0
answers
13
views
Trying to figure out issues in neural networks
I encountered a bizarre thing when using NNs to approximate a dynamic programing problem.
PROBLEM
A typical Bellman equation
$$V(s) = \max_c\{R(s,c) + \beta V(c)\}$$
$s$ and $c$ stand for the state ...
- 11
8
votes
2
answers
653
views
Feature Engineering : combine a categorical Feature and a continuous Feature
When we analyze data, we can observe several variables that may contain mutual information. For an example, there can be a binary variable such as $Y=\text{Have you ever smoked?}$. Then there will be ...
- 7,692
0
votes
1
answer
36
views
'City climate twins': similarity measure for monthly temperature data
Cities are getting hotter due to climate change. I want to communicate this intuitively through a concept of 'city climate twins': ie. which other city will my city's climate resemble several decades ...
- 104k
0
votes
0
answers
26
views
SVM kernels corresponding to different types of distance measures
This answer to Data normalization for RBF kernel points out that RBF kernel implies Eucledean distance. Are there kernels corresponding to other popular distance/dissimilarity measures, such as Bray-...
- 3,492
0
votes
1
answer
29
views
Detecting and Forecasting Intermittent Time Series
I am building a model to forecast some metrics. Those metrics are quite seasonal giving me good forecasts as shown below:
However, some new requirements dictate that I target those forecasts per ...
12
votes
2
answers
2k
views
Scaling the backward variable in HMM Baum-Welch
I am just trying to implement the scaled Baum-Welch algorithm and I have run into a problem where my backward variables, after scaling, are over the value of 1. Is this normal? After all, ...
CommunityBot
- 1
2
votes
1
answer
3k
views
What are soft classes?
When reading the help page of tf.nn.sparse_softmax_cross_entropy_with_logits, I came across the term soft classes. I read that soft classes have ground truth labels which are a probability ...
- 23
0
votes
0
answers
25
views
Clustering algorithm recommendations
So I have somewhat of a clustering problem. What you see in the image is a set of points that need to be classified or clustered in six rows. These particular points have already been classified, that ...
1
vote
1
answer
298
views
Selecting a label smoothing factor for seq2seq NMT with a massive imbalanced vocabulary
I'm training a seq2seq RNN with a vocabulary of 8192 words. This means that the typical categorical cross entropy label smoothing factor suggested in papers like 'Attention is all you need' of $0.1$ ...
- 7,692
1
vote
3
answers
2k
views
Understanding bias and variance for different models over same dataset
Consider we have 1-D data generated by a polynomial of degree 5. Which will of thhe following give higher / lower bias and higher / lower variance?
Regression with linear basis functions
Regression ...
CommunityBot
- 1
0
votes
2
answers
42
views
How do I incorporate a patient's current age into a survival model?
I have a dataset with 10,000 patients, and for each patient, I have the following information:
Biological sex (male/female)
Baseline age (age at the time the patient joined the study)
Age at the ...
- 84.3k
0
votes
2
answers
397
views
How we can compare loss and metrics evaluated on different sizes of training and validation sets?
Let N_train, N_val and N_test are the number of examples in training, validation, and test ...
CommunityBot
- 1
1
vote
0
answers
24
views
Extrapolation of expected value in multivariate regression
Consider a random variables $Y, X, Z$. Denote by $f$ a function of the expectation of $Y$: $ f(x,z) = \mathbb{E} [Y\mid X=x, Z=z]$. Define a marginal function $$g(x) = \mathbb{E} [Y\mid X=x] = \int f(...
- 175
0
votes
0
answers
31
views
Permutation Feature Importance with an ensemble of models
I would like to use Permutation Feature Importance (PFI) with a set of different models coming from a bootstrap procedure.
I was looking at the algorithm here and was wondering how it can be used when ...
- 257
0
votes
0
answers
9
views
Error in accuracy test [duplicate]
This is an update from my previous question.
I'll put my Model Development code here for your reference:
...
1
vote
0
answers
106
views
How much of an indicator can machine learning provide in detecting good versus bad market conditions (risk)?
I generally buy and sell weekly as stock prices fluctuate with volatility. I am exploring the idea of using a machine learning algorithm to consider various economic conditions (inputs) and provide ...
1
vote
1
answer
247
views
Fixed effects on machine learning models?
I have been using fixed effects (from fixest package) on different type of panel data (longitudinal) models. Mainly parametric.
I was wondering if fixed effects could be apply to machine learning ...
CommunityBot
- 1
1
vote
1
answer
275
views
Edge Detection Convolution Intuition
I was learning about convolution and how filtering helps us to detect an edge in an image;however I still cannot not understand how the convolution process in the image below does this.
I understand ...
CommunityBot
- 1
0
votes
0
answers
56
views
Design of experiment in multiple regression
Suppose we have the following model of our environment:
$\hat{y}_t = e^{dayofweekeffect} * x_{1, t}^{\beta_0} * x_{2, t}^{\beta_1}$
which we can linearize into: $log(\hat{y}_t)= dayofweekeffect + \...
- 1
-1
votes
0
answers
18
views
Self supervised learning [closed]
I want resources and code for using self supervised learning for counting small images in a picture . i already have a code for it but it is not sufficient to count small images .
2
votes
1
answer
176
views
Is batching needed for the test set?
I'm just starting to learn about CNN (convolutional neural networks). Does the test data also need to be divided into batches, similar to how it's done with the training data?
- 7,692
1
vote
0
answers
10
views
RecSys model performance stalling at 47% AUC and F1-Score. Is the problem due to ratio of users to items in my dataset?
I'm having trouble with making my validation metrics go down for the binary_crossentropy and go up for the F1-score and AUC. I've tried tuning my hyper parameters such as the number of latent features ...
2
votes
1
answer
438
views
Machine Learning for Time Series: Train and test set overlap due to lagged target as feature – problem of data leakage?
Situation:
My objective is to apply Machine Learning (for regression problems). Therefore, I have a panel dataset of time series with daily fund data from 2018-01-01...
CommunityBot
- 1
1
vote
0
answers
37
views
Regression ML model: Data Augmentation [closed]
I'm currently working on data augmentation to my regression problem, and a (possible) solution that came to my mind was to add a perturbed dataset to the original dataset, and hence double the ...
- 111
0
votes
1
answer
792
views
How to Interpret output Coefficients of Linear Support Vector Regression?
I'm looking to interpret the output from my SVR model.
I know that with SVM you can't directly interpret the coefficients of the model but that you first have to take a dot product
With that said, ...
CommunityBot
- 1
1
vote
0
answers
128
views
Stability for algorithm implies no overfitting
We let $\mu$ be a distribution on the set $Z=X \times Y$. For any $S \in Z^n$ and $i$ in $[n]$, and we define $S^{(z,i)}$ as the vector in $Z^n$ that coincides with $S$ in all entries except the $i$-...
- 345
4
votes
0
answers
40
views
Why are approaches that approximate a random forest with a single decision not more popular?
I understand that random forests yield better performance than standard decision trees, but are less interpretable, because they do not generate a single tree. In this question, several users provided ...
- 41
2
votes
2
answers
1k
views
Using ml model output to choose another ml model input
I'm dealing with a low event rate problem (e.g. credit card fraud).
I've balanced my data with SMOTE, and ran a neural net model (cross validated with recall as the measure).
However my precision (...
- 10.8k
0
votes
1
answer
237
views
How can I generate a plot of the partitions in Isolation Forests
I have seen this plot is used to indicated how anomalies are isolated via partitioning in Isolation Forests. Is there a library to automatically plot this from a dataset? The plot I want to generate ...
CommunityBot
- 1
1
vote
1
answer
980
views
Binary Classification: changing the notion of "positive label"
Perhaps more logical/philosophical rather than math question. In binary classification setting, classes are most often not symmetrical and one of them is considered to be "positive" or "success", ...
- 55.1k
1
vote
1
answer
948
views
Micro- or macro-averaged AUC for highly imbalanced data?
I have a classification problem with 3 classes. With random forest classifier I'm getting the following confusion matrix:
The micro-averaged AUC is 0.76 and the macro-averaged AUC is 0.55.
On the ...
CommunityBot
- 1
0
votes
0
answers
9
views
Prediction when Target's lag values are part of Predictors
I'm using LGBM for regression, where the Target column's lagged values (7 columns for each lag day) are also used as predictors when training the model. Absence of the 7Day lag values severely ...
- 25
0
votes
1
answer
338
views
How to properly impute values on the test set using imputer (missForest)
I'm trying to impute some missing values on my dataset $X$. So first I shuffle and split data to obatin the train set X_train and the test set ...
CommunityBot
- 1
1
vote
1
answer
570
views
Repeating the testing/training split while performing cross-validation on rf model
I'm fitting random forest regressions on my data, and using 10 K-fold cross-validation to evaluate model performance. While re-runing the cross-validation, I noticed that the results differed between ...
CommunityBot
- 1
1
vote
1
answer
300
views
When should RNNs be used, seeing as they take much longer to train?
RNNs seem to take much longer to train in most if not all cases. I assume this is because the number of operations involved in training an RNN scales not only with the number of examples being fed ...
CommunityBot
- 1
0
votes
0
answers
9
views
Optimal formula to predict product when one of the components is certain
When an item is clicked, it generates revenue equal to the cost per click (CPC). As such, revenue = I(click) x CPC. I want to select the item with highest expected ...
2
votes
2
answers
489
views
R Squared (OOB) and R Square from correlation of prediction of test set is different?
I'm using simulated data and fitting Random Forest model for regression on a training dataset. What is confusing me is that after running Random Forest, I got R Squared (OOB) equal to 0.14. But when I ...
CommunityBot
- 1
0
votes
0
answers
13
views
Training problem in CNN model for image classification
I am training Cifar-10 dataset of 32 x 32 sized coloured images for image classification.
Here is the link/source to the dataset and its description: https://www.cs.toronto.edu/~kriz/cifar.html
The ...
- 10.8k
2
votes
1
answer
248
views
Regression on imbalanced and zero-inflated data. How to deal with less frequent values?
I am implementing a regression, however my regressor has not been able to predict the least frequent counts. I've tried adjusting the hyperparameters (as you can see below), but I haven't had much ...
- 55.1k
2
votes
0
answers
33
views
Is there any function that is convex after compounded with a squared loss (besides linear ones)?
It is known that a linear function compounded with a squared loss is convex, so one can efficiently find the optimal solution when performing linear regression. Specifically, given a data point $(x,y)$...
- 55.1k
1
vote
0
answers
31
views
Variance of Influence Functions, Cross-fitting, and the Propensity Score
Following example 2 in this paper, suppose I wanted to estimate $\psi = E[E[Y|X,A=a]] $ and I had an influence function follows:
$$
IF(\psi) = \frac{A}{\pi(X)}\{Y-\mu(X)\} - \psi
$$
where $\pi(X)$ is ...
- 30.2k
2
votes
2
answers
371
views
Adjusting precision recall curve for oversampling
I built a model for a binary target using oversampled data.
The population target prevalence is 0.25.
I oversampled to 0.5 by keeping the entirety of the minority class and sampling a portion of the ...
CommunityBot
- 1
2
votes
1
answer
356
views
How to handle weighted examples in stochastic gradient descent (with mini-batches)?
Suppose I have $M$ data points $x_i$ and associated weights $w_i > 0$. I want to optimize a function,
$$F(\theta) = \frac{1}{M}\sum_i w_i f(x_i;\theta)$$
in the parameters $\theta$. I will assume ...
CommunityBot
- 1
9
votes
2
answers
612
views
Relative variable importance/explained variation from a single model fit
I am seeking a measure of relative variable importance or relative explained variation that will apply to all types of linear and nonlinear regression models and that requires only fitting one model. ...
- 83
0
votes
0
answers
12
views
How to determine the split in a decision tree?
The goal here is to grow a decision tree with depth 1 based on the small dataset below. I can use the sum of squared errors as measure to select the best variable split.
$$
\begin{array}{|c|c|r|}
\...
- 211
0
votes
0
answers
12
views
95% confidence interval for C-index after running elastic net for a cox model, and how to get net reclassification index
Can someone please show me how to get 95% confidence interval for c-index in the elastic net codes below:
...
- 2,775
0
votes
0
answers
33
views
Policy Gradients Therom for Episodic Cases: How are different formulas related?
While I am generally aware of how the Policy Gradients algorithms work theoretically, I was recently a bit confused between two definitions of the Policy Gradient and later the derivation of the ...
- 2,268
1
vote
1
answer
104
views
How do GANs handle discrete outputs?
Let's consider some fictive task of generating binary images of size 200x200 (each pixel should be either 0 or 1). As far as I understand, the generator will output 200x200 values between 0 and 1 ...
- 101
0
votes
0
answers
14
views
Dealing with Class Imbalance and Probability Calibration in My Image Recognition Model [closed]
I've implemented a classification model and followed a suggestion to address class imbalance issues. However, I'm still encountering problems with the model's performance.
This is my Model Testing ...