Questions tagged [probability]

Probability is the branch of mathematics concerning numerical descriptions of how likely an event is to occur or how likely it is that a proposition is true.

Filter by
Sorted by
Tagged with
18 votes
4 answers
2k views

XGBoost outputs tend towards the extremes

I am currently using XGBoost for risk prediction, it seems to be doing a good job in the binary classification department but the probability outputs are way off, i.e., changing the value of a feature ...
user avatar
12 votes
1 answer
8k views

Are the raw probabilities obtained from XGBoost, representative of the true underlying probabilties?

1) Is it feasible to use the raw probabilities obtained from XGBoost, e.g. probabilities obtained within the range of 0.4-0.5, as a true representation of approximately 40%-50% chance of an event ...
user avatar
  • 373
9 votes
3 answers
2k views

Loss Function for Probability Regression

I am trying to predict a probability with a neural network, but having trouble figuring out which loss function is best. Cross entropy was my first thought, but other resources always talk about it in ...
user avatar
  • 191
7 votes
2 answers
3k views

Confidence intervals for binary classification probabilities

When evaluating a trained binary classification model we often evaluate the misclassification rates, precision-recall, and AUC. However, one useful feature of classification algorithms are the ...
user avatar
  • 203
6 votes
3 answers
3k views

Why does the naive bayes algorithm make the naive assumption that features are independent to each other?

Naive Bayes is called naive because it makes the naive assumption that features have zero correlation with each other. They are independent of each other. Why does ...
user avatar
  • 1,215
6 votes
2 answers
13k views

Xgboost predict probabilities

When using the python / sklearn API of xgboost are the probabilities obtained via the predict_proba method "real probabilities" or do I have to use ...
user avatar
5 votes
1 answer
5k views

How does binary cross entropy work?

Let's say I'm trying to classify some data with logistic regression. Before passing the summed data to the logistic function (normalized in range $[0,1]$), weights must be optimized for desirable ...
user avatar
  • 389
5 votes
3 answers
9k views

How to convert an array of numbers into probability values?

I would like some help with respect to certain numerical computation. I have certain arrays which look like: Array 1: [0.81893085, 0.54768653, 0.14973508] Array 2: [0.48078357, 0.92219683, 1.02359911]...
user avatar
5 votes
3 answers
5k views

How to get probabilities values with keras?

tensorflow version = '1.12.0' keras version = '2.1.6-tf' I'm using keras with tensorflow backend. I want to get the probabilities values of the prediction. I want the probabilities to sum up to 1. ...
user avatar
  • 51
4 votes
1 answer
604 views

Compute parameters of a PDF (probability density function) for which no closed form expression is available

I would like to compute parameters such as mean, variance, quantiles, etc. for a PDF which is only given as a piece of code. That is, it can only be evaluated numerically at given points; no closed-...
user avatar
4 votes
1 answer
2k views

Data-generating probability distribution, probability distribution of a dataset, in ML

In Goodfellow I, Bengio Y, Courville A. Deep learning. MIT press; 2016 Nov 10. http://thuvien.thanglong.edu.vn:8081/dspace/bitstream/DHTL_123456789/4227/1/10.4-1.pdf p. 102 (for example), it is said ...
user avatar
4 votes
1 answer
646 views

What does a predicted probability really mean, without considering the accuracy of the underlying model?

Say I've built a (completely unrealistic) classification model in Keras that gives me 1.00 accuracy. And next, I would like to use my model on some new, unseen data, and use ...
user avatar
4 votes
2 answers
220 views

How useful is Bayesian Inference

Last few months, I had been exposed to Bayesian Inference in ML course With further investigation, I come to place where there is MCMC technique to simulate the ...
user avatar
4 votes
3 answers
145 views

Distance between very large discrete probability distributions

I have 192 countries where each country has some value for 1 million attributes which sum up to 1 (a discrete probability distribution). For any one country most of the values for the attributes are 0....
user avatar
4 votes
1 answer
262 views

Can someone please explain what this sample function is upto?

So there is a function in Dino_Name_Generator at Deeplearning.ai notebook ...
user avatar
  • 2,245
4 votes
1 answer
171 views

which loss function (if any) optimizes the calibration graph

The calibration graph is the predicted versus actual probability(see http://scikit-learn.org/stable/modules/generated/sklearn.calibration.calibration_curve.html). Is it possible to optimize the ...
user avatar
4 votes
1 answer
128 views

Relation between an underlying function and the underlying probability distribition function of data

I heard and read a lot of times the following statements and got a lot of confusion over time. Statement 1: The goal of machine learning is to get a function from the given data Statement 2: The goal ...
user avatar
  • 147
4 votes
2 answers
638 views

xgboost or lightgbm to handle Binomial problems [duplicate]

I have a dataset containing a column of trials, a column of successes and other features; and, obviously, I can generate a probability column. I would like to use gradient boosting methods (like ...
user avatar
3 votes
1 answer
2k views

What is difference between Bayesian Networks and Belief Networks?

While reading some articles about Bayesian Networks, I came across many occurrences of Belief Networks. Do both of these terms mean the same thing or is there any difference between Bayesian Networks ...
user avatar
3 votes
3 answers
454 views

What is the meaning of likelihood?

I am studying Bayes probability applied to machine learning, and I have encoutered the concept of likelihood, which I don't understand. I have seen that the Bayes rule is: $P(A|B)=\frac{P(B|A)P(A)}{...
user avatar
  • 751
3 votes
2 answers
3k views

How to use different classes of words in CountVectorizer()

Suppose I have a piece of writing and I want to assign probabilities to different genres (classes) based on its contents. For example Text #1 : Comedy 10%, Horror 50%, Romance 1% Text #2 : ...
user avatar
  • 255
3 votes
1 answer
4k views

Pytorch doing a cross entropy loss when the predictions already have probabilities

So, normally categorical cross-entropy could be applied using a cross-entropy loss function in PyTorch or by combing a logsoftmax with the negative log likelyhood function such as follows: ...
user avatar
3 votes
2 answers
1k views

Very low probability in naive Bayes classifier

I have designed NB classifier from scratch in python for binary classification problem. There are total 220 records out of which 85 records belongs to 'Yes' class and 135 to 'No' class. My classifier ...
user avatar
  • 189
3 votes
2 answers
2k views

Softmax classifier never allows for 100% probability in LSTM?

When working with LSTM I am using a softmax classifier and a one-hot encoded vector approach. The softmax looks like this: $$S(h_i) = \frac{e^{h_i}}{\sum e^{h_{total}}}$$ notice, LSTM's result is a $...
user avatar
  • 2,586
3 votes
1 answer
614 views

Confidence value for face recognition

In the context of face recognition I have the following histogram: blue bins count the comparison distances for "self matches" (comparing two images of the same person). Orange bins count the ...
user avatar
  • 131
3 votes
2 answers
166 views

Seeking advice on knowledge discovery

Background Information I work for a fire department in Florida and the fire chief posed a question to me; At any given moment in time during the calendar year 2018, how many fire trucks are busy, ...
user avatar
3 votes
2 answers
4k views

Calculating an estimate of KL Divergence using the samples drawn from distributions

Given two sets of samples drawn from two different distributions, is it computationally possible to get an estimate of KL-Divergence between the two distribution using these samples? Here I am ...
user avatar
3 votes
1 answer
497 views

Re-sampling of a Histograms Bins

I would like to be able to resample a histograms bins without having access tot he raw data. And just to be clear, by resample, I mean to change the number of bins and still provide a good estimate ...
user avatar
3 votes
1 answer
45 views

Can the 'bin size' in a histogram be thought of as a regularity constraint?

When thinking about a histogram as an estimate of the density function, is it reasonable to think of the bin size as a parameter that constrains the local structure of that function? Also, is there a ...
user avatar
  • 31
3 votes
1 answer
976 views

Selecting the right algorithm for match probability prediction

Looking for assistance kick-starting a new machine learning scenario. In this case I need to pair one entity (ex. person) with a group of entities (ex. other people) given a history of matching ...
user avatar
  • 131
3 votes
2 answers
126 views

Analysis of probability distribution of each features and Machine Learning

While I know that probability distributions are for hypothesis testing, confidence level constructions, etc. They definitely have many roles in statistical analysis. However, it is not obvious to me ...
user avatar
  • 359
3 votes
1 answer
145 views

Using softmax for multilabel classification (as per Facebook paper)

I came across this paper by some Facebook researchers where they found that using a softmax and CE loss function during training led to improved results over sigmoid + BCE. They do this by changing ...
user avatar
3 votes
0 answers
45 views

How to make machine learning model that reports ambiguity of the input?

Suppose I want to build a neural network regression model that takes one input and return one output. Here's the training data: ...
user avatar
  • 305
3 votes
1 answer
50 views

Marginalization of joint distribution

I am trying to understand how you marginalise a joint distribution. In my case I have a fair coin, $P(C) = \frac12$ and a fair dice $P(D) = \frac16$. I am told I win a prize if I flip the coin and ...
user avatar
3 votes
5 answers
2k views

How to predict the probability of an event?

I have a dataset where a set of people donated for charity along with the dates of the donation. I have to find the probability of each donor donating in the next three months. Data is available from ...
user avatar
  • 141
2 votes
1 answer
8k views

What is "noise" in observed data?

I am reading pattern Recognition and machine learning by Bishop and in the chapter about probability, "noise in the observed data" is mentioned many times. I have read on the internet that noise ...
user avatar
  • 127
2 votes
1 answer
180 views

Correct way of calculating probability

I have some data which shows how many orders were made by a certain customer group that bought a certain product type: And the same format but showing how many refunds were made: I am trying to ...
user avatar
2 votes
1 answer
97 views

why naive is needed in Naive Bayes ,what happens if naive is not included in Bayes theorem?

Im trying to understand why naive is needed in Naive Bayes and everyone says Naive Bayes assumes the input features (predictors) are not correlated hence they are not dependent on each other . i want ...
user avatar
  • 1,201
2 votes
1 answer
502 views

Classification problem approach with Python

I am a Python beginner, just getting into machine learning and need advice on the approach i should use for my problem. Here is an example of my data-set. Where the RESULT is a corresponding INDEX ...
user avatar
2 votes
3 answers
479 views

In supervised learning, what does "Estimating $p(y \vert x)$" mean?

I read chapter 5.1.3 of Joshua Bengio's deeplearning book, which says: supervised learning involve observing examples of random vectors $\textbf{x}$ and associated value or vector $\textbf{y}$ and ...
user avatar
2 votes
2 answers
4k views

Are real analysis and measure theory essential to learn for data science? [closed]

Some people say data scientists don't necessarily need to know real analysis and measure theory, but for others, real analysis and measure theory are very important for the undersdanding of kernel ...
user avatar
2 votes
1 answer
288 views

What is the difference between maximum likelihood hypothesis and maximum a posteriori hypothesis?

I am a student and I am studying machine learning. I am focusing on the concept of Bayesian learning and I have studied the maximum likelihood hypothesis and the maximum a posteriori hypothesis. I ...
user avatar
  • 751
2 votes
1 answer
119 views

Problem understanding probabilistic generative models for classification

I am a student and I am studying machine learning. I am focusing on probabilistic generative models for classification and I am having some troubles understanding this topic. In the slide of my ...
user avatar
  • 751
2 votes
2 answers
148 views

How to model user choice probability: binary model vs multi class model

Let's say Morpheus has multiple users to offer colored pills(from an infinite set of colored pills), there are in total 3 unique colored pills(red, blue, green) Morpheus can offer. The trick is, ...
user avatar
  • 73
2 votes
1 answer
396 views

Convolution and Pooling as Infinitely strong priors

I am currently reading about Convolutional Neural Networks in Deep Learning Book. I am stuck on section 9.4 titled "Convolution and Pooling as Infinitely String Priors". Could someone please ...
user avatar
2 votes
2 answers
758 views

Why does a belief network need to be represented using a directed acyclic graph (DAG)?

I would have thought that it was because DAGs preserve the dependency relationships between the variables, but I am currently unsure.
user avatar
2 votes
2 answers
611 views

A3C - Turning action probabilities into intensities

I'm experimenting with using an A3C network to learn to play old Atari video games. My network outputs a set of probabilities for each possible action (e.g. left, right, shoot), and I use this ...
user avatar
2 votes
1 answer
75 views

Generating a random numpy ndarray of 0 and 1 with a specific range of 1 values

I want to generate a random numpy ndarray of 0 and 1. I want that the number of occurrences of 1 in every specific rowto range between 2 and 5. I tried:...
user avatar
  • 51
2 votes
1 answer
89 views

Bayes theorem on the probability of an object drawn at random using percentages (not Naive Bayes)

It's the normal Bayes equation but I'm not sure if I've calculated this correctly or how to check my work, here is a somewhat similar question but I wasn't sure if our math was the same, the question ...
user avatar
2 votes
1 answer
94 views

Stacked Model performance?

I am currently working with a dataset that seems very easily separable and I have an accuracy of 99% for SVM (NN-98%, RF-98%, DT-96-97% and I have checked for leakage & overfitting). As part of my ...
user avatar
  • 63

1
2 3 4 5 6