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
1answer
15 views

A good machine learning approach for distribution of a whole?

So I had done with different classification, regression and clustering approaches for predictions of values etc. I was wondering if there is a machine learning approach for distribution of a whole ...
1
vote
1answer
10 views

Intuitive explanation of Minimum Covariance Determinant (MCD)

I am an undergrad working on Anomaly Detection on an 8 dimensional dataset, with PYOD, which relies on the MCD in the sklearn's MinCovDet. I tried reading Minimum Covariance Determinant and Extensions,...
0
votes
1answer
238 views

Multi class classification using Naive Bayes

I have components basically divided into two main categories. AWS and Azure. For eg: ...
0
votes
0answers
5 views

Three related functions to M- Estimators

I have spent a lot of time reading about Robust regression especially M- estimators and their related functions (objective function, score function, and weight function). I know that M-estimators have ...
0
votes
0answers
10 views

Generate synthetic data given AUC for multivariate X

I need to simulate data for a fixed AUC and for multivariate X. I came across the link below which explains for the univariate x. Generate synthetic data given AUC. Has anyone had any ideas or ...
0
votes
0answers
11 views

Comparing a hurdle model with a “direct” model

I am building xgboost models for prediction of insurance risk, the risk being assumed to follow a tweedie distribution with tweedie variance power between 1 and 2 (https://en.wikipedia.org/wiki/...
0
votes
1answer
2k views

How to design a many-to-many LSTM RNN in Keras

I have timeseries data with 1 minute cadence with 4 features, and I want to try to predict the time-evolution of 2 of these features using a RNN using LSTMs in Keras. My aim is to predict the e.g. ...
0
votes
1answer
20 views

How do I perform a logistic regression w/ SMOTE

I want to understand which variables lead to an infection by parasites in a tree. Hence, I want to use stepwise logistic regression based on AIC. First, I describe what I would do, and then my code ...
0
votes
0answers
17 views

Using normal distribution to do rough hyperparameter tuning

I wanted to ask if this is a valid way of doing hyperparameter tuning. I have 7 parameter for my model. Since I have too many parameters to do a grid search, I was going to try a different method: Do ...
3
votes
1answer
189 views

Generate synthetic data given AUC

I'm experimenting with ROC-AUC for binary classification problems. I want to generate synthetic data for a given AUC score. The ...
0
votes
1answer
5 views

Learning curve vs training (loss) curve?

In machine learning, there are two commonly used plots to identify overfitting. One is the learning curve, which plots the training error and test error (y-axis) over the training set size (x-axis). ...
1
vote
1answer
26 views

Days as dummy variables

How should you treat "days" as a variable from a statistical perspective. The "days" variable can be defined as an integer describing the day of the year as follows: days = 1,2,3,4....
0
votes
0answers
8 views

Xgboost parameter scale_pos_weight for cost sensitive learning

I am using XGboost for classification in Python in a very unbalanced scenario. I know that, according to documentation, I can set the scale_pos_weight, to handle ...
0
votes
1answer
32 views

is there a way to train the models in batches? i m not talking about deep learning

I have 800,000 records. The training with all the records at once is not possible. Is there a way i can train a model on 50k records then continue training on next 50k records and so on... basically i ...
0
votes
1answer
25 views

Which is the best approach for permutation testing for deep learning models?

Should I just shuffle my labels and train my model for a certain number of iterations? The result of permutation testing should give low testing/validation accuracy, right? Is it necessary to do a ...
264
votes
8answers
87k views

Why is Euclidean distance not a good metric in high dimensions?

I read that 'Euclidean distance is not a good distance in high dimensions'. I guess this statement has something to do with the curse of dimensionality, but what exactly? Besides, what is 'high ...
2
votes
1answer
313 views

How can I overfit a fully-connected neural network to predict RGB values from (x,y) coordinates?

The problem is the following: Given a single 3-channel image (e.g. 200x150), I constructed a dataset where the features are the pairs of (x,y) coordinates and the targets are the (R,G,B) values. Each {...
0
votes
0answers
6 views

When training embeddings should negative samples be distinct from the context?

Suppose I am training word2vec embeddings with skipgrams. I have defined my context and my target word, and now I am looking for negative samples. It just so happens that I randomly sample a word that ...
11
votes
3answers
2k views

Compare classifiers based on AUROC or accuracy?

I have a binary classification problem and I experiment different classifiers on it: I want to compare the classifiers. which one is a better measure AUC or accuracy? And why? ...
1
vote
0answers
10 views

Pruning in Decision Trees?

Pruning is a technique in machine learning and search algorithms that reduces the size of decision trees by removing sections of the tree that provide little power to classify instances. I know what ...
2
votes
0answers
10 views

Reconstruction Error: Principal component analysis vs Probabilistic prinicpal component analysis

I am working through the book "Machine Learning: A Probabilistic Perspective". After introducing PCA and Probabilistic PCA, the following graphic is shown (the upper two graphics ...
0
votes
0answers
3 views

heuristics for gamma in rbf kernel

My question is a follow-up to this question: SVM rbf kernel - heuristic method for estimating gamma. Basically, I want to find interesting values for gamma by first calculating the pairwise distance ...
0
votes
0answers
6 views

Classification model produces extremely low test accuracy, although training and validation accuracies are good for multiclass classification

I'm trying to do alphabet classification for American Sign Language. So it's multiclass classification task with 26 classes. My CNN model gave 84% training accuracy and 91% validation accuracy, yet ...
1
vote
1answer
1k views

scale factor for MAD for non-normal distribution

I understand that the scale factor for normally distributed data is 1.4826 to convert it to a pseudo standard deviation like quantity which could be used with the median for determining confidence ...
3
votes
2answers
39 views

What is my model learning- Linear vs Non-Linear

I am trying to validate if my understanding of the topic is right. Linear Models pick up on a single representation of a class (read can only pick one) whilst a Non-Linear Model like a Neural Network ...
2
votes
1answer
262 views

Choice of an evaluation metric for a graph clustering algorithm

I have instances for which the only thing I know is 70% of the distance matrix. I know some of these points form groups of correlated points (each point of a group is "close" to every point of the ...
0
votes
0answers
14 views

Why don't companies use on-device ML to serve ads to protect privacy? [migrated]

Eg: The ad inventory can be stored in the cloud. Using on device ML, a process then matches the ad to the user in order to serve the best ad to the right user. Are there any limitations to this? (I ...
1
vote
1answer
29 views

Limitations of K-Means Clustering [duplicate]

I was going through a document of Western Michigan University to understand the limitations of K-means clustering algorithms. Below is the link: https://cs.wmich.edu/alfuqaha/summer14/cs6530/lectures/...
1
vote
1answer
28 views

Bias and Variance of a Decision Tree for Classification

There are lot of discussion about bagging and boosting in the context of decision trees and how Random Forest and other methods helps to tackle bias and variance. But how exactly can I measure bias ...
0
votes
0answers
10 views

Datascience Project steps [closed]

I have completed "An Introduction to Statistical Learning: With Applications in R". I am familiar with most of the concepts but I'm not sure in which order do I need to perform them. When do ...
1
vote
0answers
8 views

Performance metric when cost of predicting Positive point as negative is very high, mainly in medical domain

I encountered following question and its answer but I do not know the justification. Original Question: Given an imbalanced data, 90% -ve and 10% +ve. When the cost of wrongly predicting Positive ...
1
vote
1answer
824 views

CNN for a regression problem

I have tons of matrices as inputs and their corresponding outputs, which are also matrices. In other words, my goal is to train a network that would predict me the matrix output based on the matrix ...
0
votes
0answers
18 views

Logistic regression: dependent observations

I have a time series dataset. One series contains ~200 data points. Each series describes one sports match. I have converted each series into ~200 samples such that each sample contains somehow ...
2
votes
1answer
311 views

machine learning techniques for classifying images with timestamps

I'm doing 2-class image classification (determining whether an object is present or absent in images) with CNNs. The dataset is a bunch of photos with continuous timestamps. And I observed that ...
0
votes
0answers
13 views

Can one make meaningful predictions from “censored data”?

In an ML project, I'm using patients' clinical and demographic features to predict their treatment outcomes. Among over 300K records, only 6,022 are known to have experienced severe conditions (ICU, ...
1
vote
2answers
2k views

Alternating least square formula

So I was reading about the alternating least square algorithm used for movie recommendations. Let's say that: X - user ratings Y - movie ratings The result is computed by this formula: In the ...
0
votes
0answers
6 views

How to test for inter-individual differences using JASP

I am trying to test if interindividual differences can mediate a relationship between my IV and DV. How can I do this?
0
votes
0answers
4 views

Is there a metric for the gains curve?

as we do have a ROC AUC metric, I was wondering if there is a specific metric to evaluate the gain curve?
2
votes
0answers
20 views

PCA for non linear manifolds - Yann Lecun, Deep Learning Course, Question from a Lecture

I was watching the following lecture and at the very end of it, one of the students asked LeCun about using PCA for expression and pose feature extraction. https://www.youtube.com/watch?v=0bMe_vCZo30&...
0
votes
0answers
5 views

What is a class variable with a dominant mode?

This is my first time posting a question on such type of platforms. I hope, someone will be willing to explain it to me:) So here is the question: What is a class variable with a dominant (and ...
2
votes
2answers
228 views

Learning the Confidence of a Neural Network

Suppose I want to train a deep neural network for classification. The network takes an input vector $x$, and maps this to an output vector $y$. Now, $x$ is of length $n$ and is in fact composed of a ...
0
votes
0answers
8 views

When might test error be lower than training error

In Ex. 2.9 of Elements of Statistical Learning (2nd edition), it asked to prove that for the ordinary least squares estimate, the expected MSE in the training set is $\leq$ than that in the testing ...
0
votes
0answers
18 views

Which specific type of recurrent neural network (RNN) is required to solve this supervised sequence problem?

I have a variety of features and an sequence target variable of fixed dimension 1x20 e.g. [0,0,0,1,0,...]. I've been reading up on the potential of RNNs to predict sequences. What specific methods/...
2
votes
1answer
157 views

What is the difference between Generative Adversarial Networks (GAN) and Generative Antagonistic System (GAS)?

What is the difference between Generative Adversarial Networks (GAN) and Generative Antagonistic System (GAS) in the neural network?
0
votes
0answers
11 views

How to correctly weight observations in a decision tree

I'm building a boosting model and trying to fit a decision tree for the weak learner with a set of observation weights. I've seen two ways to do this: 1) bootstrap sample so that you have a higher ...
10
votes
1answer
3k views

The variance-covariance matrix of the least squares parameter estimation

I'm learning Linear Regression for Regression from "The Elements of Statistical Learning". Why The variance-covariance matrix of the least squares parameter estimates is easily derived from (3.6) ...
0
votes
0answers
7 views

How to split heavily imbalanced datasets and obtain solid results for a finished model?

I've just finished my first paper that required evaluating the performance of some algorithms on a specific dataset. And after reading and understanding the basics, like how to clean your data and ...
0
votes
1answer
136 views

Prediction for a large number of discrete numbers other than classification and regression

I am dealing with a problem where the output of my model, can have numbers like 1-3000 (around) (score in a game). This is like a score in a game. Giving a least squared error setting, for a model, ...
0
votes
1answer
18 views

GLM coefficients from caret - Machine Learning

I have a quick question I can't seem to find a good answer to, so I hope this makes sense and please let me know of any important information I may leave out. I've been using the machine learning ...
0
votes
0answers
12 views

Linear regression with missing label? Which is the best model to be applied? [closed]

I do have a dataset related to a ticket resolution system with different ticket categories. I am asked to predict the -time to resolution- for each ticket being this value available in the training ...

1
2 3 4 5
305