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]

Methods and principles of building "computer systems that try to automatically improve with experience."

0
votes
0answers
5 views

Will merging class labels in pre-trained model hurt transfer learning?

I'm using a large image dataset labeled with 15 classes to train a ConvNet model. The resulting model will then be used to enable transfer learning in a tiny dataset labeled with only 3 classes. The 3 ...
0
votes
0answers
19 views

Multivariate time series with a binary dependent variable

I am currently working on a multivariate time series data set with 1 dependent variable (y) and 60 independent variables ($x_1$,$x_2$,....$x_{60}$). The dependent variable is a binary variable (0 or 1)...
0
votes
0answers
9 views

Determine input array that approximates a target output array from complex numerical simulation?

I believe the following problem is ideally suited to a machine intelligence approach, but am unsure where to start. I've used scikit-learn previously with downloaded datasets, but the following seems ...
0
votes
0answers
21 views

Does this mean my model is useless?

Long story short, I have a random forest I've created. It's mean absolute error is .209 for the test set. The (scaled) standard deviation of the y column is .201 (for all the y column data, not just ...
-1
votes
0answers
16 views

Understanding MLE for a Gaussian Naive Bayes classifier

I am trying to develop a text classifier and I'm reading about MLE to help me understand the process. I came across this example: and I wanted to try this myself. I'm running into a problem and so ...
1
vote
0answers
15 views

How do PGMs factor in to modern ML?

I just finished the three-part series of Probabilistic Graphical Models courses from Stanford over on Coursera. I got in to them because I realized there is a certain class of problem for which the ...
0
votes
0answers
24 views

ML Regressor model performance conclusion RSME vs STD DEV

Perhaps my question is still slightly silly but apparently even though lot of folks talk about how to evaluate the rightness of your model there still blur the right evaluation procedure at least for ...
0
votes
0answers
15 views

How do I know how to combine kernels in e.g. Gaussian Process Regressions

Looking for optimal parameters usually is a pain for most Machine Learning tasks. In most cases, however, we can perform a grid search to find out about which parameters do the job better or worse. ...
0
votes
0answers
8 views

Hierarchical vs multi-class, when to use what?

I am looking for suggestions from machine learning research perspectives about using hierarchical classifier vs multi-class classifier. For example, if I have to classify 3 classes, let's say, 2 ...
1
vote
2answers
41 views

How to check randomness in a machine learning dataset?

given a standard machine learning dataset, is it possible to check whether the relationship between the inputs and outputs is random or not? If the relationship between inputs and outputs is random (...
1
vote
0answers
23 views

Averaging coefficients obtained from linear models of infinite subset in regression

If we estimate coefficients for a linear model by repeating experiments infinite number of times (Which means collecting samples or datasets infinite number of times and fitting a linear model for ...
0
votes
1answer
58 views

why vc-dim of the range space (R^m, {lines in R^m}) is 2?

I need to prove that for any positive integer $m \in N$ the range space $(X , Y)$ where $X = \mathbb{R^m}$, and $Y$ consists of lines in $\mathbb{R^m}$ has VC dimension 2. but I know that the VC ...
1
vote
0answers
61 views

What is the most intuitive proof that Gaussian kernel is positive definite?

I have general form of Gaussian kernel $K(x,x')=\exp(-\|x-x'\|^{2})$ (just not considering $\sigma$). I tried to prove its positive definiteness via Gram matrix properties, but couldn't. Is there any ...
1
vote
0answers
8 views

Uplift modelling

I have been trying to build a uplift model which gives incremental probability of a customer responding to a treatment. I am thinking of using pylift library for my model, I had few questions ...
0
votes
0answers
9 views

Program synthesis - control flow

How do we reason about whether a neural network can learn a control flow structure? For example, Given inputs a, b, c, I have a program: ...
3
votes
0answers
28 views

Quantifying uncertainty of predictions for new data in the regression tree

I used Regression Learner to train my data. I held out 25% of the input for validation and ran different models for training. Based on the results using RMSE and R-squared, I decided to go for the ...
1
vote
1answer
43 views

Feature Interaction Strength in Catboost

I was wondering if anyone knew how the feature interaction strength is calculated in the catboost package. The documentation https://catboost.ai/docs/concepts/output-data_feature-analysis_feature-...
2
votes
4answers
118 views

Using Trend as a feature in time series sliding window?

I have a time series, and i am using overlapping sliding window to extract features from each window and label it accordingly. In this Overlapping window of size n, i want to extract trend (linear fit)...
1
vote
1answer
18 views

Given two features, one a string and other a categorical, what are the encoding rules?

I have two features in my dataset I'm using to help predict a binary outcome. Based on my features, I'm trying to figure out which I need to drop a dummy to avoid the dummy trap. One feature is a ...
1
vote
0answers
27 views

Random Forest Regression with sparse data in Python

I am working on a Random Forest regression model to predict housing prices. I have about 500k rows of data with the following information: 1.House area in square meters. 2.Number of rooms. 3.City. ...
0
votes
1answer
26 views

A standard name for a formula to “Maximize true positives while minimize false positives”

I am using an evaluation metric to reward the true positives and penalize the false positive ones retrieved by a function $f(\cdot)$. Indeed, it can be represented as follows: $\frac{\texttt{|TP|} - \...
1
vote
1answer
14 views

Reference point in projection axis of SVD (singular value decomposition)

I am watching a YouTube video on SVD, and attempting to recreate some of its examples to better understand the internal machinery of the algorithm. In one of the slides, the instructor mentions that ...
1
vote
1answer
25 views

Recommended machine learning algorithms for a 10-class image classification problem with only 1900 samples

I am trying to determine the right approach to take for an image classification problem which involves 10 classes and only 1900 samples. The images (1288 x 964 resolution) are of industrial parts ...
0
votes
0answers
16 views

Simple Logistic regression with keras from 4 features

I am trying to create a simple NN using keras, i have data in this form: which contains rows of 4 numbers ranging from 0 to 100, based on these values i am trying to predict the outcome as 0 or 1 ...
1
vote
1answer
47 views

What is exact 'learning' on a deep learning model?

This is something I do not understand. Consider a regular Machine Learning model. I have a lot of pictures of cats and dogs and I feed them to the model and train. I am new to machine learning but I ...
4
votes
1answer
23 views

Convergence to gradient in limit of variance

I came across this equation in the original GAN paper (pg 2 https://papers.nips.cc/paper/5423-generative-adversarial-nets.pdf): $$\lim_{\sigma \rightarrow 0} \nabla_{\bf x} \mathbb{E}_{\epsilon \sim \...
0
votes
0answers
21 views

How do Gradient Boosted Trees calculate errors in classification? [duplicate]

I understand how gradient boosting works for regression when we build the next model on the residual error of the previous model - if we use for example linear regression then it will be the residual ...
2
votes
1answer
30 views

Modelling approach - tennis match predictions

I am working with a dataset about a fictitious type of sport which is fairly similar to tennis: One has to win 5 points to win a game, 4 games to win a set and 3 sets to win the match. However, there ...
0
votes
1answer
69 views

Why do Deep learning models need larger data sets compared with classical ML

It might be a very basic question but i cant find a good answer for that. Why do deep learning networks need larger data set then what needed for more classical ML models such as SVM , classical ...
0
votes
0answers
18 views

Using a priori knowledge in a classification task

I'm working on a classification task, related with text classification, where texts to be classified are requests for technical support, and the classes are technical guys which issues can be assigned ...
-1
votes
1answer
7 views

How to Quantize Vectors using Kmeans?

I have a bunch of entities, with each instance having 40 features, so a 40-dimensional object. I cluster them using K-means. Now, I need to quantize them. I want to ask two questions: How to ...
1
vote
0answers
17 views

Adaboost - Show that adjusting weights brings error of current iteration to 0.5 [closed]

I'm trying to solve the following problem but I've gotten sort of stuck. So for adaboost, $err_t = \frac{\sum_{i=1}^{N}w_i \Pi (h_t(x^{(i)}) \neq t^{(i)})}{\sum_{i=1}^{N}w_i}$ and $\alpha_t = \frac{...
1
vote
0answers
24 views

How are these two expected values equal? [closed]

From understanding machine learning: How are the two expected values equal in the proof of the theorem in the red box below? Notation: $D$ is a distribution on the set $Z$, $E_{z \text{ ~ }D}$ ...
1
vote
1answer
16 views

Is it necessary to constrain the size of the neighborhood in LLE to be less than the space dimensionality?

The wikipedia entry on Locally Linear Embedding (LLE) says that LLE can be broken into stages, the first of which is to learn a barycentric linear model of the data with its $k$-nearest neighbors: $...
0
votes
0answers
23 views

CNN: Relationship between kernel size and node size in convolution layer

I have a question and that is maybe because I have a misunderstanding. In CNN, the convolution layer uses different filters (kernels). Let's imagine I have a network with 81 nodes in the input layer (...
2
votes
1answer
22 views

Natural Language Processing: Basic Dimension Reduction with SVD of a Co-Occurence Matrix

Given sentences I enjoy flying. I like NLP. I like deep learning We can form a Co-Occurrence Matrix as follows: Now we can apply Singular Value Decomposition to this matrix to get $X = U \Sigma V^...
0
votes
0answers
12 views

Incremental Probability apart from the propensity model

I have built a model which predicts if a customer is going to pay given some kind of intervention, say calls. The model is used to generate propensity scores for each customer and then the calling ...
2
votes
2answers
103 views

Why is the risk equal to the empirical risk when taking the expectation over the samples?

From Understanding Machine Learning: From theory to algorithms: Let $S$ be a set of $m$ samples from a set $Z$ and $w^*$ be an arbitrary vector. Then $\Bbb E_{S \text{ ~ } D^m}[L_S(w^*)] = L_D(w^*)...
1
vote
0answers
28 views

What is an autoregressive model - terminology with respect to machine learning

In Wikipedia, an autoregressive model is defined in terms of an AR(p) linear process as The autoregressive model specifies that the output variable depends linearly on its own previous values ...
1
vote
1answer
35 views

Machine learning model underperformance on unseen data

This is a follow-up question to a question I had previously posted on this forum We conducted an experiment on 100 subjects and obtained a dataset that was used to train a machine learning model that ...
2
votes
1answer
82 views

bias variance tradeoff — properties that do not follow

Going through this lecture note on bias-variance trade-off, I didn't follow the latter part of this paragraph. It shows the common situation in practice that (1) for simple models, the bias ...
0
votes
1answer
33 views

How to simulate discrete data from a Poisson distribution? [closed]

I want to use R to simulate discrete data with missing values from a Poisson distribution. I have tried this: simdata<-rpois(1000,2) but when I checked if there ...
3
votes
0answers
32 views

Require understanding regarding the concept of restricted estimators

I was reading "The Elements of Statistical Learning Book by Jerome H. Friedman, Robert Tibshirani, and Trevor Hastie" where I encountered the following: The part tells us that the RSS criterion will ...
1
vote
0answers
18 views

Practical realities of updating a trained model with new data

In my day to day work, I train models on data using R packages that have no extension for Bayesian priors. I will generally have a large dataset to start off with, and add new data as needed. Any ...
3
votes
1answer
70 views

Predict when a user logins next

I am building a user login prediction system. This is my first time building any prediction system. Main aim is to predict when a user might login next in future. That is i need to predict "Time". I ...
1
vote
1answer
31 views

What is the baseline of the F1 score for a binary classifier?

I know how to calculate the baseline for the accuracy of a binary classification problem: I simply always predict the majority class, e.g. if there is 94% True values and 6% False values, my baseline ...
12
votes
2answers
772 views

Is decision threshold a hyperparameter in logistic regression?

Predicted classes from (binary) logistic regression are determined by using a threshold on the class membership probabilities generated by the model. As I understand it, typically 0.5 is used by ...
0
votes
0answers
18 views

How to use Factors from Exploratory Factor analysis in further analysis?

I have performed an exploratory factor analysis on a large data set as a dimension reduction technique. I have come up with 20 factors that group together my predictor variables. However, I am not ...
0
votes
1answer
23 views

How do ConvNets self-organise to have a hierarchical segmentation of higher- and lower-level features?

As far as I know, each layer of a convolutional neural network used for image classification specializes in recognizing a different part of an image. At earlier stages in the network, more rudimentary ...
1
vote
0answers
24 views

Supervised classification with different row length dataset

I need to make some predictions based on some situations: the problem is how to enconde situations themself. A row in the dataset will represent a single situation but here it comes the problem. A ...