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.
19,969
questions
0
votes
0
answers
2
views
Why detections count is not equal to unique truth count in YOLOv4 test result report?
I trained model using YOLOv4 on GPU, CUDNN and openCV (python) with AlexeyAB\Darknet with multi-label on windows environment. These labels are 25 classes (from 0 to 24). Then I test the model and I ...
- 1
2
votes
0
answers
19
views
Multiclass proper scoring rule decomposition: (weighted) average across the categories?
I have found a Python function that calculates the decomposition of various proper scoring rule, such as Brier score and log loss. However, it does not seem to accept arrays as arguments, so if I want ...
- 58k
1
vote
0
answers
23
views
Which book help us to do the best EDA for data science? [closed]
I am new at data science. I would like to make ML models in orde to make predictions, but I am lost when I have to make interpretations to data... I would really like to understand the data with ...
2
votes
0
answers
24
views
Analogue of landscape conjecture in likelihood theory or Bayes?
The so-called landscape conjecture in machine learning says that in high dimensions, most critical points of the loss surface are saddle points rather than poor local minima.
Out of curiosity I was ...
- 951
0
votes
0
answers
9
views
Is There a Standard Metric for Evaluating Treatment Impact Considering Action Cost in Uplift Models?
I'm currently exploring Uplift modeling, specifically the use of the Conditional Average Treatment Effect (CATE) metric:
$$ \tau(t', t, x) := \mathbb{E}[Y | X=x, T=t'] - \mathbb{E}[Y | X=x, T=t] $$
...
- 115
2
votes
1
answer
93
views
How to interpret the deviance plot by boosting models
This plot is taken from a gradient boosting regression example in the scikit-learn documentation.
What does deviance mean? How should this plot be interpreted? In which case do we have over/...
- 55
0
votes
0
answers
27
views
How to interpret error histogram and scatter plot?
I'm new to machine learning and I'm having a hard time interpreting how good my model is performing. I have run my model on a dataset of wine attributes trying to predict alcohol content based on the ...
- 133
0
votes
1
answer
27
views
Relation between number of classes, their separability and classification difficulty
Intuitively, it seems that a classification problem with more classes is "harder" than the same problem with fewer classes. However, this also seems to depends on the separability of the ...
- 1
3
votes
2
answers
172
views
How to combine two linear models?
I am trying to predict housing prices using as few variables as possible. One way that yielded the best results so far is splitting the data into two data-sets ($houses < 100m^2$ and $houses >= ...
- 31
-3
votes
0
answers
342
views
Python: Nonlinear Poisson Regression with two independent variables, or something similar like a Random Network? [closed]
I'm looking for a algorithm to use Nonlinear Regression with Poisson for predictive purposes in Python or something similar anyway. I would need something similar to Poisson regression, but with some ...
- 115
0
votes
0
answers
27
views
Loss stops dropping in a convolutional classification network [closed]
I'm training my first CNN, but loss stops at 1, and its plot also shows spikes:
This is the code:
...
- 1
4
votes
1
answer
141
views
Regression coefficient on a triangle using geometry
I am encountering a question as follows:
Let $X, Y$ be two independent uniform random variable on $(0,1)$. We consider the regression model $Y = \beta_1 X + \beta_0$, given the restriction that $X + Y ...
- 143
0
votes
0
answers
10
views
Machine Learning College predictor [closed]
I have a training data set containing College names,student rank, branch, college cutoff. Which prediction model should I use to predict the list of colleges a student will get admission in according ...
1
vote
0
answers
17
views
Given conjugate prior and posterior distributions, what is the PRIOR predictive distribution? [closed]
I am doing an assignment on my statistics class. We had 1 lecture about bayesian parameter estimation, where we were taught about the following formula (and it's discrete form, if $h(\theta)$ was ...
- 111
4
votes
0
answers
73
views
As Brier Score = MSE, does MSE in a regression have a calibration-discrimination decomposition?
When the outcome of a supervised learning problem is binary and probabilities are predicted, Brier score can be decomposed into a measure of calibration and a measure of discrimination.
...
- 58k
-1
votes
0
answers
5
views
Does Faster RCNN have an upper bound on the number of bounding boxes which it can predict in a single image? [closed]
Asked today
Modified today
Viewed 9 times
Report this ad
Closed. This question is not about programming or software development. It is not currently accepting answers.
This question does not appear to ...
2
votes
1
answer
66
views
Could classical astronomic discoveries be made using ML?
Imagine that prior to the time of Copernicus, scientists could run machine learning algorithms (somehow).
Would it be possible to discover the truth of the geocentric model with their use?
Would the ...
- 511
2
votes
0
answers
16
views
Looking for a way to train a model to learn optimal parameters/hyperparameters of clustering
I have 5000 docs, each is a review. For each review, I'm plotting the sentences in a semantic dimension. Now, I'm applying clustering to these points for each review. The success of my model depends ...
- 21
1
vote
1
answer
67
views
Regularization Problem and Reproducing Kernel Hilbert Space
The following shows part of the page 169 of The Element of Statistical Learning that I want to make clear.
We have
$$\min_{f \in \mathcal H_K}[\sum_{i = 1}^NL(y_i, f(x_i)) + \lambda\Vert f\Vert_{\...
- 187
-1
votes
0
answers
9
views
What type of Autoencoder is this?
I have an autoencoder structure which is given below. Now, I want to know what type of autoencoder is this? Is it Bayesian or Gaussian?? Or is it something else?
...
- 1
0
votes
0
answers
8
views
Time Series Prediction for Variable-Length Input with Fixed Sampling Period
I'm working on a time series prediction task where I need to predict the parameters (amp, phase, freq, offset) that best fit a variable-length input series. The samples are always at the same period.
...
1
vote
1
answer
27
views
Relative Risk or Odds ratio using machine learning
I am thinking of using machine learning type classification models to compare them with the traditional approach of logistic regression (dichotomous outcome) in a sample of patients with diabetic foot ...
- 57
-1
votes
0
answers
10
views
so the question is about calculating MAP and PPD. I know the formulas for both, but find it confusing, so can someone explain step by step? [closed]
Now suppose that you run each model, and they make the following predictions:
p(yt+1 | yt, θ1) = .4 p(yt+1 | yt, θ2) = .75 p(yt+1 | yt, θ3) = .6.
What is the maximum a posterior estimate p(yt+1 | yt,...
0
votes
0
answers
40
views
Bert Used for generative AI
I have a doubt regarding using "Bert" as a generative model. I know Bert can be used for classification or fine-tuning the question-answering. However, is it possible to use Bert to generate ...
- 175
0
votes
0
answers
23
views
Is the right set of customers to use in my churn model?
I am creating a churn model in python. The full dataset has around 90k records stretching back many years. I'm using a subset of the full dataset. This subset only includes clients where we've worked ...
- 127
1
vote
1
answer
22
views
How to predict outcome with zeros, and both negative values and positive values in ML model?
My Outcome is quarterly sales, which is right_Skewed and the range of it is large b/c different groups purchased differently. For Example, there are lots of groups that made zero or less than 10 ...
- 11
3
votes
0
answers
73
views
Is Brier score strictly proper in multi-label problems?
In problems where one of $3+$ categories can be observed and we prodict the probability of each category being observed, it is known that the Brier score is a strictly proper scoring rule that is ...
- 58k
0
votes
0
answers
23
views
Bias-Variance tradeoff proof [duplicate]
why are y_hat and ε independent?
I think ε dose effect y_hat
0
votes
0
answers
14
views
Does Positional Interpolation Change Llama's Architecture?
I'm currently exploring Meta's positional interpolation method, which aims to increase the context size in their large language model. This method extends the context length from n x n into n′ x n′. ...
1
vote
1
answer
117
views
Help with Classification model for S&P500
I have started a project in order to develop my coding skills, where I am predicting next month's S&P500 return direction based on some macroeconomic and financial variables. These datasets have ...
- 13
0
votes
0
answers
15
views
what does edge mean in the context of sample space?
This is a follow-up question on this post
It uses a term 'edge'. For example, it says
I understand that extrapolation is harder than interpolation. And I understand that if we choose a point to ...
0
votes
0
answers
10
views
Should unprocessed features be kept in the dataset along engineered ones?
I'm working on a Machine Learning classification problem that has five Service Spending features (among many others) in its dataset, each sample is a customer. For instance, here are the first three ...
- 1
0
votes
0
answers
18
views
Reporting internal validation from .632 bootstrap in caret
I am developing Machine learning prediction models using the caret package in r (Elastic net, SVM, Random Forest, XGBoost). I have 650 cases with 104 having the event of interest. Instead of splitting ...
- 1
0
votes
0
answers
61
views
How do I proof or disprove that the size of a document influences classification accuracy?
I have a multinomial Naive Bayes model that classifies a document into 1 of 178 classes. This goes well about 50% of the time, so it's pretty good (for my use case) considering the large set of ...
0
votes
0
answers
13
views
Conditional Average Treatment Effects: What is the logic underlying using the covariates to predict the residualized scores?
I've consulted a few references, including this online book, and I've got a good grasp on using double machine learning to estimate ATE after debiasing/denoising the treatment variable (T) and the ...
1
vote
0
answers
27
views
What kind of machine learning model could I use on this dataset?
I am a beginner to data science. I found this dataset that covers natural disaster incidents in Afghanistan from 2016 - present. Here are the 13 columns: REGION (South West, North, etc), PROV_CODE (...
- 11
2
votes
3
answers
77
views
Neural network training: Early stopping vs. reducing the number of neurons/layers
Assuming we have a regression task (1D-output, values between 0.0 and 1.0) with 100 input dimensions and a classical MLP with ...
- 179
0
votes
0
answers
11
views
Occam's Razor: Commit to a hypothesis before vs. commit after seeing the data
In the section on Occam's Razor of the book "Understanding Machine Learning, S. Ben David et al." (its free online version here), the authors wrote:
if we commit to any hypothesis before ...
- 478
0
votes
0
answers
19
views
Optimizing Customer Request Classification: Handling Multilabel Notes and Varied Description Lengths
I'm working on a project to classify customer requests based on a dataset that includes request descriptions, labels and sentiments. The goal is to predict the label (and sentiments) of customer ...
- 1
1
vote
1
answer
51
views
Is model distillation an ill-defined problem?
Model distillation (or knowledge distillation) consist in making a student model learn from a teacher model in order to eventually use the student model as an alternative to the original teacher model....
- 11
3
votes
1
answer
137
views
Unpack the notation used in Wikipedia's decomposition of the Brier score
Wikipedia has an article about the Brier score whose notation confuses me.
The article starts out easy enough by defining the Brier score to be:
$$
BS = \dfrac{1}{N}\overset{N}{\underset{i = 1}{\sum}}\...
- 58k
-1
votes
0
answers
32
views
Computationally effective way to ensure the neural network output with a mean of 1
I want to enforce the output of a feedforward neural network to have a mean of 1. I understand that I can use softmax function or do some postprocessing (i.e., divide the output by its mean). However, ...
- 9
0
votes
0
answers
9
views
tree plotted by ctree has no splits
I've computed a ctree model, but the individual tree plotted is just one level, with all samples (below). How could this scenario have occurred? My response variable is a principal component and the ...
- 1
2
votes
3
answers
69
views
How does neural networks create a boundary in case of more than 3 classes?
I understand that a neural network is a universal function approximator, meaning that it can learn to approximate any function that maps inputs to outputs. For a classification task, it creates a ...
- 53
0
votes
0
answers
4
views
What type of rotational invariance information can I get from interest points (coordinates) of corners from an image?
Assume that you are using the FAST algoritm for corner/feature detection. You pick an image and run the FAST algorithm.
Question:
All these green dots are actully coordinates in x and y direction. ...
- 381
4
votes
0
answers
74
views
Can the calibration-discrimination decomposition of Brier score be viewed as the bias-variance decomposition of mean squared error?
The mean squared error has a famous decomposition into bias and variance.
$$
\text{MSE} = \text{bias}^2 + \text{var}
$$
Brier score is also a mean squared error calculation, and Brier score has a ...
- 58k
0
votes
0
answers
7
views
Tree structured data prediction
I have tree-structured dataset about parents and childs values. Each data node (except father-node) has some float values. I need to predict childs values.
For being more obvious, I made a picture to ...
- 101
1
vote
1
answer
94
views
Multivariate Time Series dataset preparation
I am a bit confused with the time series dataset preparation. From the internet, I saw all examples which used tree-based models, had input features and target defined as:
...
- 61
0
votes
1
answer
74
views
What's the best clustering algorithm for Fraud Data?
Background
I'm working on a Fraud dataset with 500,000 samples, and 130 features.
There are:
98 numerical features,
32 categorical features,
There are missing values in:
7 numerical features,
12 ...
- 399
0
votes
0
answers
27
views
Hyperparameter selection after nested cross-validation and making comparisons with DeLong's test
I have already read all the associated questions on the topic but couldn't find a clear answer. I initially split my data into training (80%) and hold-out testing (20%). Then, I am performing nested ...
