Stack Exchange Network

Stack Exchange network consists of 174 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 [gaussian]

The tag has no usage guidance.

0
votes
0answers
9 views

VAE: Calculating reconstruction loss when the true posterior is continuous

In Python / Tensorflow, I am currently struggling with how to compute the VAE reconstruction loss $ E_{q_\phi(z|x^{(i)})} [log p_{\theta} (x^{(i)}|z)] $ (which is equivalent to the negative cross ...
0
votes
0answers
18 views

Feature selection in RTLS based on BLE [on hold]

I need to build a machine learning model for RTLS(Real Time Location System) based on BLE beacons & BLE gateways. I have 3 BLE gateways which is placed across the floor and I have recorded the ...
-1
votes
1answer
50 views

Can I call this graph as a gaussian? [closed]

My program is a chatbot. It has rule to represent the state that user is talking to the bot at node level n. I have 1 to 9 nodes in the application. Here is the ...
0
votes
1answer
25 views

Limits of using a normal distribution in Bayesian inference

When applying a Bayesian inference method such as Gaussian Process Regression (GPR), the assumption of a prior and likelihood function following a normal distribution is inherent. One can use an ...
0
votes
1answer
187 views

Gaussian distribution in python without using libraries

I am implementing Gaussian distribution of a variable, but it gives multiple bell shapes. It should be a single bell shape. Below is my code and plot. ...
0
votes
0answers
18 views

How to find number of Gaussians in GaussianMixture?

I have been using the elbow method to find optimal K value in K-means. Now while using GaussianMixture (Using Spark), how to find the number of gaussians say 2 OR 3 OR more...? Thanks
1
vote
1answer
55 views

Probabilistic Outlier Detection (edited + clarified)

Measured data $D \in \mathbb{R}^3$, every $d^i \in D$ is $d^i_{(x)}$, where the $x=[x_1, x_2]$. Simply said, the measured data are function of $x$. It is known, the dependency is linear, such as: $$...
2
votes
0answers
20 views

What are the reasons for drawing initial neural network weights from a Gaussian?

Are there theoretical or empirical reasons for drawing initial weights of a multilayer perceptron from a Gaussian rather than from, say, a Cauchy distribution?
2
votes
1answer
59 views

What mu and sigma vector really mean in VAE?

In standard autoencoder, we encode data to bottleneck, then decode with using initial input as output to compute loss. We do activate matrix multiplication all over the network and if we are good, ...
1
vote
0answers
14 views

Parameter of Conditional Gaussian Distribution

I'd like to understand how to determine the parameter of conditional gaussian distribution. Following is the network architecture of VUNET which learns the conditional gaussian distribution $q(z|x, \...
0
votes
0answers
15 views

How to assign prior probabilities while using Gaussian Process bandits?

I implemented the work based on Srinivas, "Gaussian Process Optimization in the Bandit Setting: No Regret and Experimental Design" and it looks like my code is working. My problem is that when the ...
1
vote
0answers
37 views

Why are observation probabilities modelled as Gaussian distributions in HMM?

HMM is a statistical model with unobserved (i.e. hidden) states used for recognition algorithms (speech, handwriting, gesture, ...). What distinguishes DHMM form CHMM is the transition probability ...
1
vote
0answers
20 views

Training Gaussian Restricted Boltzmann Machines with Noisy Rectified (nrelu or ssu) linear hidden units

I'm not sure how to implement this architecture. I'm following this thesis (pages 17-19) or this paper but I'm not sure how to train it. I want to use this to extract features from raw audio. I know ...
3
votes
1answer
24 views

Mixture Density Network: determine the parameters of each Gaussian component

I am reading Bishop's Mixture Density Network paper at: https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/bishop-ncrg-94-004.pdf This is a good paper, but I am still confused about ...
1
vote
1answer
25 views

Where can we find the application of bayes's theorem in Bayesian optimiation with gaussian processing

I am trying to learn bayesian optimisation by following this tutorial. However, until now I don't get the relation between bayes's theorem to the gaussian process formalism. Any ideas?
0
votes
0answers
14 views

Fitting Gaussian process to set of distributions (mean values + variances)

So I am trying to use Gaussian processes to model human function learning (in a reinforcement learning -ish setting). Humans are trying to guess the value of some stimulus based on the feedback they ...
1
vote
1answer
68 views

Understanding GMM-MMI

While studying Gaussian mixture models and the expectation maximization algorithm, I also came across a number of studies that used 'Discriminative training' for the models that addressed some ...
1
vote
2answers
78 views

How to tune bandwidth in machine learning kernel model?

Gaussian kernel $k(x,y) = \exp(-\lVert x-y \rVert^2/\sigma^2)$ has a hyperparameter $\sigma$. I know grid search cross validation, but this would require a lot of computation since computational ...
1
vote
0answers
53 views

Gaussian Process Regression: does feature normalization affect final log-likelihood of model?

I'm trying to learn a Gaussian Process Regressor in SKLearn. I tried it both with and without feature (and output) normalization, and even though results seem similar-ish, the reported log marginal ...
2
votes
0answers
90 views

Wasserstein distance between Gaussian and the empirical distribution

Wasserstein distance between two gaussians has a closed form solution. Does the same hold for the distance between a Gaussian with a fixed variance(say 1) and the empirical data distribution? ...
0
votes
0answers
27 views

NewbieQuestion: Predicting with noisy input

I am beginner, working on a problem with an aim to predict a variable which has some noise added to it. I just need to know if I am on the right path. Here is what I tried. I found that the noise ...
1
vote
0answers
57 views

results from “Google Vizier: A Service for Black-Box Optimization”

In Google Vizier: A Service for Black-Box Optimization it is show what kind of optimizers google uses internally. Figure 6 shows some benchmark results compared to Random Search. Am I right in ...
0
votes
0answers
74 views

SKlearn: Gaussian Process Regression not learning

I'm trying to fit a GP using a GaussianProcessRegressor, and I notice that my hyperparameters are still at their initial values. I did some stepping in gpr.py, but ...
0
votes
0answers
33 views

getting prediction as array like [0. 0. 0. 0. 0. … 0. 0. 0.] even accuracy in test data is above 70%

I am using sklearn's preprovided classifiers for training ,my data, i have used bag of words model to make data. I have used technique TF-IDF to normalize my data. The classifier i used is mainly :- ...
0
votes
0answers
86 views

Gaussian model calibration with PyMC3

I am attempting to implement bayesian model calibration under the classical Kennedy-O'Hagan framework using PyMC3. Briefly, the setup is as follows: I have $n$ ...
0
votes
0answers
16 views

The difference in variable correlation maximum limit

From the jupyter notebook the max limit of the variable corr is 0.95 as shown in that code sample below: @interact_manual(var1 = (1,9), var2 = (1,9), corr=(-0.95,0.95,0.05)) In the interact display ...
0
votes
0answers
30 views

Intuition for prediction based on SVM trained with Gaussian kernel

I'm currently taking my first steps in machine learning using Andrew Ng's Coursera course. The follow Octave/MatLab script optimizes a SVM using a given kernel function on training set X,y: svmTrain ...
0
votes
0answers
32 views

How does a zero mean vector improve Feature extraction in Autoencoders

I've been studying a paper where a generalized Bayesian Stacked Denoising Auto-encoder is being used for feature extraction, my concern is how does taking zero as the mean vector help . here is the ...
1
vote
1answer
102 views

How do i use the Gaussian function with a Naive Bayes Classifier?

I would like some help with the maths and to check I have understood the algorithm correctly. So I’ve been learning off of a video and I tried my own example. At the end of it I got dodgy results; I ...
1
vote
2answers
1k views

Why does PCA assume Gaussian Distribution?

From Jon Shlens's A Tutorial on Principal Component Analysis - version 1, page 7, section 4.5, II: The formalism of sufficient statistics captures the notion that the mean and the variance ...
2
votes
2answers
590 views

Working with Data which is not Normal/Gaussian

What happens if my data/feature is not normal? Can I still use machine learning algorithms to utilize such data for predictions? I noticed in many data sciences courses, there is always a strong ...
-3
votes
1answer
80 views

Which Normal distribution a point belongs to?

I have estimated normal distributions for two classes (0 and 1). The distributions for the two classes are, \begin{align} X_0 &\sim \mathcal{N}(\mu=549.96,~~ \sigma=549.96) \\ X_1 &\sim \...
2
votes
1answer
182 views

Replacing missing value by class conditional mean

I have two classes, $p(x|y=0)$ and $p(x|y=1)$ with ${{\mu }_{0}}$ and ${{\mu }_{1}}$ as mean and shared covariance matrix $\Sigma $. Now, I have a missing feature ${{x}_{n}}$ for a particular ...
1
vote
0answers
277 views

ValueError ill-defined empirical covariance when running a Gaussian Mixture Model

I am getting the following error when running a Gaussian Mixture Model: ...
1
vote
1answer
466 views

GausianNB: Could not convert string to float: 'Thu Apr 16 23:58:58 2015'

I'm beginner in python so please bare with me. I'm trying to solve one machine learning problem using GaussianNB. I've certain fields which are not in proper date format, so I converted it into UNIX ...
1
vote
1answer
69 views

Classification followed by regression to handle response variable that is usually zero

I have a data set consisting of a bunch of predictors (mostly unbounded or positive real numbers) and a single response variable that I wish to predict. The response is typically exactly zero -- ...
0
votes
1answer
176 views

How are the positions of the output nodes determined in the Kohonen - Self Organizing Maps algorithm?

In the Cooperative stage of Kohonen's SOM, the neighborhood for a winning neuron(output node). In most cases, the neighborhood function happens to be the Gaussian Function. For example, $$h_j,_i = exp(...
4
votes
1answer
168 views

Inductive Bias in Gaussian process

All supervised learning techniques have some kind of an inductive bias. What is the inductive bias in Gaussian process models ?
-1
votes
3answers
429 views

How to check if a data is in gaussian distribution in R or excel?

I know about the fitdist() function from the fitdistrplus package in R, however, I am not able to use it to predict a gaussian ...
3
votes
1answer
4k views

Transform a skewed distribution into a Gaussian distribution

I have a skewed distribution that looks like this: How can I transform it to a Gaussian distribution? The values represent ranks, so modifying the values does not cause information loss as long as ...
1
vote
1answer
87 views

EM parameter estimation for conditional Gaussians

Let $$X_1\sim N(\mu_{X_1},\sigma_{X_2}^2)$$ $$X_2\sim N(\mu_{X_2}, \sigma_{X_2}^2)$$ where $\mu_{X_2}=c+aX_1$. Also, I have data $D$ (with missing values on $X_1,X_2$). How can I update/estimate the ...
3
votes
1answer
203 views

Gaussian Mixture Models EM algorithm use average log likelihood to test convergence

I was investigating scikit-learn's implementation of the EM algorithm for fitting Gaussian Mixture Models and I was wondering how they did come up with using the average log likelihood instead of the ...