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
6 views

Are feature importances from tree based models directly actionable for business?

If my response variable say is "has_repurchased" [0 or 1] and I have all customer level features. Can I rank the features in order of importance from the random forest model and report them as whats ...
0
votes
0answers
2 views

Modeling both linear and non-linear relationship

Apology for being verbose and all the typos or mistakes. This problem has been bothering me for a while and I really hope you can help me with it. Let's say I want to model quarterly sales for a ...
1
vote
0answers
4 views

Asymmetric or unequal misclassification costs in random forest

I have a general question about asymmetric costs. In machine learning problems, there are times when the cost of a false positive is different from the cost of a false negative. Accordingly, models ...
0
votes
0answers
11 views

Features that are important according to random forest are not significant when logit model was used. How to interpret?

I have a feature set for each customer [age, gender, income, lifestyle, & so on...] and a response variable say: has_repurchased. I use a logit model summary which shows income & gender to ...
0
votes
0answers
9 views

How to derive a ranking function from observations

I have a employee dataset with the following 5 details. projects completed customer ratings number of bugs reported customer complaints profit I want to rank ...
1
vote
0answers
21 views

How does regularized regression overcome the p > n problem?

So, I understand why simple linear or logistic regression will have infinite solutions in this case (good answers here and here). But while LASSO will only select n features, Elastic net does not have ...
0
votes
2answers
16 views

Question from Machine Learning Textbook on Number of Models

In the James, Witten, et.al Statistical Learning textbook, it says the following: "Unfortunately, there are a total of $2^p$ models that contain subsets of $p$ variables." Can someone please ...
1
vote
1answer
374 views

What is the difference between accuracy and AUC score which one to trust?

I've 300 samples with multiclass classification problem with 3-classes. I implemented SVM in R programming. Below is the output which I am really confused. Can anyone logically explain to me what is ...
3
votes
1answer
762 views

Hoeffding's Inequality

I am studying the feasibility of learning from the book Learning from Data. The author uses a bin analogy to discuss the feasibility of learning in a probabilistic sense. I have certain questions to ...
0
votes
0answers
6 views

High accuracy on both training and validation but very low on test set

My CNN model has about 96~97% accuracy on both training and validation sets. But when submitting the test set it got only 24% accuracy. Here's my model: ...
36
votes
4answers
28k views

How does rectilinear activation function solve the vanishing gradient problem in neural networks?

I found rectified linear unit (ReLU) praised at several places as a solution to the vanishing gradient problem for neural networks. That is, one uses max(0,x) as activation function. When the ...
10
votes
5answers
8k views

Why f beta score define beta like that?

This is the F beta score: $$F_\beta = (1 + \beta^2) \cdot \frac{\mathrm{precision} \cdot \mathrm{recall}}{(\beta^2 \cdot \mathrm{precision}) + \mathrm{recall}}$$ The Wikipedia article states that $F_\...
1
vote
1answer
28 views

Inputting playing card values to aneural network

I am trying to create a NN to play a card game wherein each state is represented by the hands of 4 players. Every round, the hand of each player is decreased by 1 (discarded). Each player starts with ...
1
vote
1answer
24 views
5
votes
1answer
332 views

Unsupervised pre-training for Reinforcement Learning

Since the advent of many unsupervised learning methods, as a pretraining step for the main supervised task (mostly under the name of Deep Learning), it shouldn't be strange to ask, what is the current ...
1
vote
0answers
8 views

Recognizing whether a written and spoken number is the same

For our ML assignment we have three datasets. The challenge is about checking whether a written and spoken number refer to the same number. We're using the MNIST dataset with handwritten numbers, and ...
4
votes
1answer
4k views

LSTM cell output activation for series

This question is asked from the perspective of finding out if there's a more efficient way for an LSTM to act more as a regression entity rather than just assigning only probabilities to the next ...
2
votes
2answers
394 views

Proof of correctness of normal equation

I was taking an online course and saw linear regression being by gradient descent The intuition behind why the method would work seemed plausible. I tried understanding normal equation as to why ...
1
vote
1answer
444 views

Influence of word counts from DTM on LDA with Gibbs Sampling

I'm trying to wrap my head around Topic Modeling based on LDA with Gibbs sampling (Griffiths, Steyvers 2004: Finding Scientific Topics). What struck me when reading some Python implementations like ...
1
vote
2answers
51 views

Finding Relationship between Categorical and Continuous data

A subset of my dataset looks as follows where cells in "cat1_ids" column contains list of "cat1" categories and cells in "person_id_list" column contains list of persons id. There are 2000 "cat1" ...
0
votes
3answers
386 views

What algorithm should I use to predict a continuous dependent variable from multiple continuous & categorical independent variables?

I'm software engineer of an E-commerce company, facing a problem like this: An e-commerce shop sells their products daily and wants to know what conditions that might improve their sales. I'm ...
0
votes
0answers
13 views

Multivriate Time Series Model - ARIMAX

I have Weekly Units sales data of products for 2 years (104 weeks). And I am trying to forecast the Unit sales for each productid for next 8 weeks.. Please find the data image below. note: Productid ...
0
votes
0answers
4 views

Estimating a changing transit time between inputs and output

I work with a chemical process in which there is a time lag between the inputs (raw material quality and cooking parameters) and the output (final product quality). The problem is that the time lag ...
0
votes
1answer
11 views

Calculating a continuous variable with regression trees

I have sample records with several attributes (predictors) and a predicted variable Yes/No. What I need is, given new data that omits the column Yes/No, to know what is the probability of Yes. Note ...
1
vote
0answers
21 views

What is the relation between a loss function and an energy function?

A loss function is a function that measures the distance between the expected value and the actual value of a model (an example of a loss function is the cross entropy). An energy function can be ...
0
votes
1answer
7 views

Setting bias of output layer for imbalanced datasets

From a blog post from Andrej Karpathy on training neural networks: Initialize the final layer weights correctly. E.g. if you are regressing some values that have a mean of 50 then initialize the ...
0
votes
1answer
14 views

Training Perceptrons with Backprop

Is it possible to train a simple perceptron with a threshold activation function such as this one: https://en.wikipedia.org/wiki/Perceptron with Backpropagation instead of the perceptron rule? is it ...
0
votes
0answers
16 views

Cannot seem to converge beyond a loss of 3 on an object detector being trained on YOLO

Data The you only look once YOLO algorithm is used for object detection. I have scoured the internet for resources on how to tackle this problem, but there seems to be a lot of resources that point ...
0
votes
1answer
45 views

What is the difference between policy-based, on-policy, value-based, off-policy, model-free and model-based?

I'm trying to clear things out for myself, there are a lot of different categorizations within RL. Some people talk about: On-policy & Off-Policy Model-based & Model-free Model-based, Policy-...
0
votes
1answer
27 views

Why test and training dataset should have same number of columns(variables)?

Why can't i train the model with specific number of varaibles and test it with more or less variables. (i know i will get error when i do this). But what is the resaon behind this? The main concept ...
0
votes
0answers
11 views

How to generate a project plan template from a list of project plans?

Consider the data for 3 users from the same domain: “Design the UI”, “Develop/code the UI” and “Discuss changes with the client” are the most common tasks. The duration could be a simple average. So,...
0
votes
0answers
20 views

Why we use calibration of Machine Learning models?

From the different websites I got to know how to perform calibration of models. But can anyone tell me the reason behind performing calibration of machine learning models?
1
vote
0answers
26 views

Do unbiased regression coefficents yield better prediction?

I ask myself if a have a omitted variables bias in my regression modell the coefficients of the model are biased so the mse growth because this coefficents are biased right? So does it mean if i ...
3
votes
2answers
339 views

Exploiting features in a multiarmed bandit scenario

I am facing a challenging problem: Say I have shirts of three different colors (same price). And say I am running a strange kind of store in into which people come in one by one, and I can show ...
0
votes
0answers
26 views

Calculating the number of neurons and the number of hidden layers for a neural network MATHEMATICALLY

I have a fair idea that a lot of research has been done and is still underway to explore the science behind the black art of a neural network (NN) architecture, i.e., accurately calculating the number ...
1
vote
1answer
178 views

Random forest permutation test: Is permutation of the training set appropriate?

I have a rather high-dimensional data set (p > 1000) with several variables ranking significantly higher than the rest in terms of variable importance (measured by Gini impurity). However, these ...
2
votes
1answer
1k views

How to penalize a regression loss function to account for correctness on the sign of the prediction?

I am dealing with a regression problem (my targets could potentially take values between -inf to +inf). To optimise my model, I have two objectives: 1) Predictions should be close to the targets. 2)...
0
votes
0answers
5 views

Datasets for Document Classification problem [on hold]

I am doing a project to make a application that can take pdf and docx documents as input and classify them into various categories such as - Financial - Government and Political - Sports and ...
0
votes
0answers
9 views

KNN works in `class` but not `caret` (Too many ties) [on hold]

I am making a KNN algorithm to predict close_price with about 80,000 rows of this data. ...
0
votes
0answers
16 views

What does the intercept represent in a model matrix?

I am making a KNN algorithm to predict close_price with about 80,000 rows of this data. ...
0
votes
0answers
22 views

relation among loss function / MLE / Bayesian estimation

I have read a lot of stuff on the relation between minimizing a loss function / maximizing the likelihood / choose a centrality measure of the posterior (Bayesian estimation); but I cannot see a clear ...
2
votes
1answer
179 views

classification on imbalanced dataset via random forest: results vary with random seed

I have a highly imbalanced dataset of about 8000 observations, with 11 features and one binary target variable. I want to predict the target labels, considering that the "1" target label occurs for 1....
0
votes
1answer
32 views

What are general practises used to divide the data into training / dev and test set?

Example: I have am building a dog vs cat classifier and I have collected data from 15 countries. Europe: 1. UK 2. France 3. Germany 4. Italy 5. Finland Asia: 1. India 2. China 3. Japan 4. Russia 5. ...
0
votes
0answers
16 views

Haul's Correlation-based Feature Selection (CFS) formula spread

I want to use the Correlation-based Feature Selection (CFS) proposed by Haul. I found this formula where $r_{zc}$ is the correlation between the summed components and the outside variable, $k$ is ...
0
votes
0answers
21 views

Is Stochastic Gradient Descent sensitive to training permutation?

I've recently read that SGD (Stochastic Gradient Descent) is one of the most popular techniques for training Machine Learning algorithms, including DNNs (deep neural networks). However, my ...
1
vote
0answers
12 views

What is parameter chaining in CNN?

I have read an article about the benefits of CNN. One of the points was "parameter chaining". What does it mean? And how does it make CNN more convenient?
1
vote
1answer
29 views

Predict song genre using LSTM

I have a dataset of songs based on genres. For example, a song may hold {5, 2, 3} as scores set for Sentimental, Rock and Jazz. In total there are 800 songs sequentially arranged. I want to predict ...
0
votes
0answers
12 views

Is the prediction with and without mean normalisation different in Collaborative Filtering?

In case of Collaborative Filtering: Given an output matrix I wish to learn parameters $\Theta$ (Parameter Vector) and X (Feature Vector). Now if I mean normalise the output matrix the values of $\...
0
votes
1answer
28 views

How can I visualize and correctly interpret the multi-output of a random forest regression?

I have multiple inputs and multiple outputs for a model I am trying to build using SkiLearn's Random Forest Regression in Python. I have imputed the missing data, divided the columns where the Ys ...
3
votes
1answer
2k views

Is the Laplace/Lidstone smoothing parameter (talking about Multinomial/Bernoulli Naive Bayes) related to the particular structure of the dataset?

I'm working with Multinomial and Bernoulli Naive Bayes implementation of scikit-learn (python) for text classification. I'm using the 20_newsgroups dataset. From the scikit documentation we have: <...