Skip to main content

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
2 votes
1 answer
32 views

Are there strategies for measuring accuracy of Euclidean distance-based similarity without ground truthing?

I have subjects with about 200 features each. These feature vectors are stored in a vector database, where similarity searching with Euclidean distance is used to find subjects that are similar to a ...
T_d's user avatar
  • 23
0 votes
0 answers
19 views

Why is my accuracy fluctuating for a while and then stuck? [duplicate]

I am building a cnn classifying model to predict images over 3 classes. The data is balanced, with 10.5k images for train ( 3.5k for each ), 3k validation images ( ...
Dragos123's user avatar
3 votes
1 answer
25 views

What is "clall" in index.Gap in "clusterSim" R package?

I am using the "clusterSim" package in my project (https://cran.r-project.org/web/packages/clusterSim/clusterSim.pdf, page 39) and I do not understand the meaning of the "clall" ...
user2702's user avatar
1 vote
1 answer
18 views

Input on methodology for analysing publicly available medical datasets

I'm being asked to analyse a publicly available medical dataset of adverse drug reactions. My data quality is quite sparse to begin with, so I've been trying my best to extract something meaningful ...
Mman231000's user avatar
0 votes
1 answer
64 views

Z-score and standard error in linear regression

I am reading Elements of statistical learning, and in the chapter on linear regression, I cannot understand the following: We have estimated the regression parameters $\beta_1, ..., \beta_p$ from $N$ ...
ge0rg's user avatar
  • 13
0 votes
0 answers
7 views

How to find a linear decision boundary of a linearly separable problem with unlimited class evaluations?

I have a binary classification problem, where my goal is to find a linear decision boundary (which I assume exists). The context of the problem is that I have an iterative optimization process, where ...
oskar0711's user avatar
3 votes
1 answer
69 views

What probability distribution is learned in this specific case? [duplicate]

I keep reading papers and blogposts where the training of a neural network is defined as learning some underlying probability distribution of the data. Imagine that you write CNN that outputs whether ...
Mah Neh's user avatar
  • 174
0 votes
0 answers
25 views

Avoiding Information Leakage in Backtesting with CPCV-Tuned Hyperparameters

I'm using Combinatorial Purged Cross-Validation to tune hyperparameters for a binary classification model applied in a month-end trading strategy. I have 6 months of data and used CPCV with 15 splits ...
June's user avatar
  • 1
2 votes
0 answers
39 views

Mathematical Introduction to Theory of Time Series Analysis

Assume that the reader has strong background in stochastic calculus (including and beyond continuous time stochastic processes like martingales and Markov chains and others, the construction of Levy ...
3 votes
0 answers
32 views

What are the "tricks" in machine learning? [closed]

I have come across a few different "tricks" in machine learning methodology, which I list below along with my rudimental understandings. The Kernel Trick: This is used in Support Vector ...
camhsdoc's user avatar
  • 409
0 votes
0 answers
24 views

Can we use convnets to learn the masked letter in a word?

I'm interested in training a CNN to learn the relationship between a word with a single masked letter and that masked letter. For example, if my model is $M$ and the input is "he-lo", it ...
Spencer Gibson's user avatar
1 vote
1 answer
121 views

Non-linear kernel for classifying data points corresponding to two concentric circles [closed]

Have seen article, while doing self-study, on Non-linearly seperable problems, here. The images as given there are here, and here. It deals a common text-book problem, where the data points are in two ...
jiten's user avatar
  • 113
0 votes
0 answers
15 views

Generalised Jensen-Shannon Divergence - Unequal Length Probability Distributions

I'd like to implement a generalized Jensen-Shannon divergence (GJSD) style test comparing 3 different probability distributions. In this respect, I looked at the Philentropy library in R with the ...
EB3112's user avatar
  • 244
3 votes
1 answer
100 views

Interpretation of a decision tree plot

For a paper, I am training different models and using LIME to simplify the blackbox models into a transparent decision tree model that I can visualize with view(tree, "mode", "graph&...
Tino D's user avatar
  • 268
9 votes
2 answers
569 views

Pearson correlation as a metric for the quality of regression models

A paper I saw used the Pearson correlation together with MSE to measure the performance of a machine learning model. After doing some research, I have seen that using Pearson correlations to evaluate ...
Degers's user avatar
  • 113
0 votes
1 answer
39 views

My model is giving inconsistent results

"I ran a Polynomial multiple regression model on a dataset with just 98 points. when I ran the model on different subsets of data (training and testing). It gives me r2 value ranging from ...
NEERAJ YADAV 's user avatar
0 votes
0 answers
17 views

Prove Decreasing Cross Entropy of outputs with Decreasing KL divergence of inputs

I am trying to prove the inequality $H(gt, y) > H(gt, y_1) > H(gt, y_2)$, given that $D_{KL}(x, x_1) > D_{KL}(x_1, x_2)$, where $y = f(x)$, $gt$ - ground truth, $D_{KL}$ - KL divergence, $H$ ...
user412867's user avatar
3 votes
1 answer
124 views

Gamma regression with XGBoost

I'll try to be brief. I have two questions about what exactly happens when I train a gradient boosted ensemble of trees using, say, XGBoost in order to perform a Gamma regression. I apologize in ...
user412834's user avatar
0 votes
0 answers
11 views

"ROC AUC reflects the likelihood that a random positive instance will be located to the right of a random negative instance". How come? [duplicate]

According to this webpage, ROC AUC reflects the likelihood that a random positive (red) instance will be located to the right of a random negative (gray) instance. Would you please explain this ...
Evan Aad's user avatar
  • 1,443
1 vote
0 answers
12 views

Problems using custom dataset using Minirocket classification [closed]

I'm working on a bigger school project, trying to classify timeseries measurements with Minirocket/Rocket. My trainingdata consists of a 1D matrix containing the measurements, and a seperate 1D matrix ...
Michael's user avatar
  • 11
4 votes
5 answers
161 views

Is it valid to exhaustively test all possible combinations of features to find the best combination?

I have about 1000 labelled observations from about 50 subjects responding physiologically under different situations and am trying to classify the situation (usually into three classes of roughly ...
user1596274's user avatar
0 votes
0 answers
23 views

Why are the prediction intervals for my DNN regression model horizontal lines?

I am working on developing prediction intervals for deep reinforcement learning. Therefore, I am following the instructions given over here. I ran a small example using a simple deep-learning model to ...
desert_ranger's user avatar
0 votes
0 answers
11 views

What to do with features causing data drift?

I dispose of labeled train data and unlabeled test data. I want to tune and validate a classifier on train data in such a way that it can have good performance on test. By conducting some ...
Yann's user avatar
  • 43
0 votes
0 answers
14 views

How is $K_p$ parameterized?

How is $K_p$ parameterized? Can you help me derive it, I am a newbie. This is my derivation, but it is different from the formula on the picture, can you help me?
user avatar
1 vote
0 answers
36 views

Predicting continuous variable based on curve

I have a dataset of a set of curves measured at different frequencies, so it is composed of curves as the figure below for example. My dataset has many more curves of course. A curve is associated ...
raygozag's user avatar
  • 613
1 vote
3 answers
226 views

Why do ML models learn probability distributions and why does it matter?

I'm aware that this question is quite silly, but I've done reading and coded NNs for quite a while now, studied backpropagation and so on. However, I don't think I ever understood what is the ...
Mah Neh's user avatar
  • 174
0 votes
0 answers
27 views

Why does $L_1$ weight decay negatively impact adversarial training for logistic regression?

I am reading through the original paper on the Fast Gradient Sign Method (Explaining and Harnessing Adversarial Examples by Ian Goodfellow et. al). We define an adversarial example as: $$\tilde{x}=x+\...
hkj447's user avatar
  • 447
0 votes
0 answers
27 views

How can I integrate time in my Implicit Feedback dataset?

I'm working on a recommendation system based on Collaborative Filtering. Specifically, I've been looking at models such as NCF (Neural Collaborative Filtering) and SAR (Simple Algorithm for ...
umus's user avatar
  • 1
1 vote
1 answer
38 views

Reason for high MSE and negative R square value

I am getting really high MSE and negative R square value. Dataset: https://docs.google.com/spreadsheets/d/1moTZS_LgOn6d74NC44i9lVcWchj-abVx/edit?usp=sharing&ouid=100514649347129021200&rtpof=...
user avatar
0 votes
2 answers
85 views

Can I use Tanh before softmax?

I am researching the use of neural networks for binary classification tasks of financial data. The output result is two-dimensional, such as [[0.5,0.5], [0.1,09]]. In the case of only 2000 small ...
Mar7's user avatar
  • 1
2 votes
1 answer
32 views

How to interpret the results of a classifier when train/test method gives much better results than cross validated one?

I need your help to understand a situation where using train and test set produces perfect results (in terms of accuracy, precision, and recall) but when cross validation is used, the accuracy on ...
letdatado's user avatar
  • 367
0 votes
0 answers
26 views

Is it possible to train a neural network to feed into a Random Forest Classifier or any other type of classifier like XGBoost or Decision Tree?

I want to create a model architecture to predict future stock price movement as such: The Goal of this model is to predict if the price will go UP or DOWN within the next 3 months. I have tried a few ...
Evank's user avatar
  • 1
0 votes
0 answers
7 views

model (constraint) sale probability of properties

I am trying to model the probability of a sale of a (vacation rental) property depending and price, property features (quality, size etc.), space and time. I am doing this at daily grain and achieve ...
cs0815's user avatar
  • 2,245
3 votes
1 answer
45 views

How to deal with extremely small training dataset in machine learning? [duplicate]

I've around 100 rows of data with labels ...
zZzZ's user avatar
  • 79
1 vote
1 answer
143 views

How to deal with extremely small training data? [closed]

I've around 100 rows of data with labels ...
zZzZ's user avatar
  • 79
0 votes
0 answers
17 views

How to improve prediction quality of LSTM model

I am trying to train an LSTM-based model in MATLAB to predict 365 next values given 365 previous values of a time series. Input shape is (1000, 365) and output shape is (1000, 365) i.e. there are 1000 ...
Avnish Bansal's user avatar
0 votes
0 answers
55 views

Which statistical model will be best for this data?

I'm trying to identify the relationship between the dependent variable and the independent variables. I've utilized linear regression, but I'm not sure if it's suitable given the distribution of my ...
Chemokine1's user avatar
2 votes
1 answer
80 views

Compute R Squared by Fixing Betas for Multi Linear Regression without Intercept results in a large R Square

I want to fix the betas in multi linear regression based on some data I have, which leads to a RSquare value less than 0% and greater than 100 % based on the projection approach mentioned in ...
godimedia's user avatar
  • 123
4 votes
2 answers
534 views

Overfitting in randomForest model in R, WHY?

I am trying to train a Random Forest model in R for sentiment analysis. The model works with tf-idf matrix and learns from it how to classify a review, in positive or negative. Positive ones are ...
Anisa's user avatar
  • 43
3 votes
1 answer
59 views

How do I develop prediction intervals for Reinforcement Learning?

I recently learned about the concept of prediction intervals (for regression) and I would like to apply them to my Deep Reinforcement Learning algorithm. I am working with a Model-Free RL algorithm ...
desert_ranger's user avatar
0 votes
0 answers
14 views

EWMA formula for SGD with momentum different than generic EWMA formula

I am currently trying to understand how SGD with momentum works, what I understand is it uses the Exponential Weighted Moving Average concept to make the updates smoother. We take weighted average of ...
learnToCode's user avatar
1 vote
0 answers
62 views

An error occurred when using the xgboost as a classifier for hiclass [closed]

Bellow it's my example when using the xgboost classifier for hiclass. My question is specifically directed to the hiClass Python package for hierarchical classification. I would like to model the ...
Ramzy's user avatar
  • 21
0 votes
0 answers
12 views

Can anyone explain me what does ACF and PCF tells from this figure?

I want to know what does it mean? Is my data stationary or not because the p value tells its stationary and what should be the order ARIMA model(p,1,q) is it (p,d,0) or (0,d,q)
theunknown's user avatar
0 votes
0 answers
31 views

Can a Catboost oblivious tree split on the same feature more than once

When training an oblivious decision tree in Catboost, can it use the same feature more than once for splitting? Let's say there is a feature age. If the first split ...
NotProbable's user avatar
0 votes
0 answers
24 views

Choosing spatial resampling functions for species distribution modeling study with different pseudo-absence sampling methods

I am conducting a species distribution modeling study using machine learning models. Since I only have presence data, I have employed various pseudo-absence sampling methods, including random ...
Marine Régis's user avatar
0 votes
0 answers
8 views

Activation functions in Neural Cellular Automata

NCAs aren't necessarily the exact same as neural networks but I am trying to do my own implementation to understand better. Do activation functions (especially in this case) serve only for training or ...
ra111's user avatar
  • 1
0 votes
0 answers
20 views

How do I interpret a Random Forest Survival C-index value relative to the Requested performance error?

I'm doing a random forest survival analysis for a school project and I'm confused about the C-index output that I'm getting relative to the Requested performance error. Shouldn't my C-index get higher ...
Jake S's user avatar
  • 41
1 vote
0 answers
10 views

PCA time series Index in R: Score Plot Doesn't Match Expected Index

I'm working on creating a PCA Index in R to understand how it works. I've used 'make up' data for this purpose. However, when I plot the scores of the first component, the results are not as expected. ...
searchandprint's user avatar
0 votes
0 answers
17 views

Strange Variance Term for Normal Prior $w^2\sigma^2$

I've attached two screenshots, one with the question and one with the answer. It seems to me that the prior is wrong and it should include $w^2$ not $w^2\sigma^2$ I apologise for, including such a ...
CormJack's user avatar
  • 161
1 vote
0 answers
39 views

Violation of i.i.d assumption in time series modeling

In time series forecasting,let's say you have $x_1, x_2, x_3, \cdots, x_t$ and the goal is to predict the the value of $x_{t+1}$ given values at previous times $1,\cdots,t$. Let's assume that the ...
Quqnus's user avatar
  • 11