Questions tagged [probability-distribution]

For questions related to AI theory that relies on the knowledge of a distribution of probabilities across one or more dimensions affecting probability. Such a distribution may be in discrete buckets, such as quartile, octile, or percentile conventions or continuous functions based on some closed form (algebraic formula). Distributions of probability are key in planning, natural language handling, and other AI objectives.

Filter by
Sorted by
Tagged with
3
votes
0answers
23 views

Difference between a distribution model and a sampling environment in Reinforcement Learning

The book from Sutton and Barto define a model in Reinforcement Learning as "something that mimics the behavior of the environment, or more generally, that allows inferences to be made about how ...
0
votes
0answers
28 views

Predicting the probability of a periodically happening event occurring at a given time

I have encountered this problem on how to predict the probability of a periodically happening event occurring at a given time. For example, we have an event called being_an_undergrad. There are many ...
3
votes
1answer
58 views

Why do we sample vectors from a standard normal distribution for the generator?

I am new to GANs. I noticed that everybody generates a random vector (usually 100 dimensional) from a standard normal distribution $N(0, 1)$. My question is: why? Why don't they sample these vectors ...
0
votes
0answers
15 views

Is there an effective way of obtaining the topic distribution for a given document from a VAE-LDA?

Is there an effective way of obtaining the topic distribution for a given document from a Variational AutoEncoder Latent Dirichlet Allocation (VAE-LDA)? Most existing public VAE-LDA codebases seem to ...
1
vote
1answer
23 views

Why are CNN binary classifier output probability distributions often skewed?

I've been working on a lot of simple resnet18 binary classifiers lately and I've started to notice that the probability distributions are often skewed one way or the other. This figure shows one such ...
4
votes
1answer
168 views

How can a probability density value be used for the likelihood calculation?

Consider our parametric model $p_\theta$ for an underlying probabilistic distribution $p_{data}$. Now, the likelihood of an observation $x$ is generally defined as $L(\theta|x) = p_{\theta}(x)$. The ...
2
votes
1answer
56 views

What is emperical distribution in MLE?

I was reading the book Deep Learning by Ian Goodfellow. I had a doubt in the Maximum likelihood estimation section (Pg 131). I understand till the Eq 5.58 which describes what is being maximized in ...
6
votes
2answers
201 views

Why is KL divergence used so often in Machine Learning?

The KL Divergence is quite easy to compute in closed form for simple distributions -such as Gaussians- but has some not-very-nice properties. For example, it is not symmetrical (thus it is not a ...
3
votes
1answer
68 views

Is this referring to the true underlying distribution, or the distribution of our sample?

I am currently studying the paper Learning and Evaluating Classifiers under Sample Selection Bias by Bianca Zadrozny. In the introduction, the author says the following: One of the most common ...
1
vote
2answers
82 views

When should one prefer using Total Variational Divergence over KL divergence in RL

In RL, both the KL divergence (DKL) and Total variational divergence (DTV) are used to measure the distance between two policies. I'm most familiar with using DKL as an early stopping metric during ...
1
vote
0answers
41 views

Estimating $\sigma_i$ according to maximum likelihood method

Let be a Bayesian multivariate normal distribution classifier with distinct covariance matrices for each class and isotropic, i.e. with equal values over the entire diagonal and zero otherwise, $\...
0
votes
1answer
46 views

How to calculate v min and v max for C51 DQN

Background: In C51 DQNs you must specify a v-min/max to be used during training. The way this is generally done is you take the max score possible for the game and set that to v-max, then v-min is ...
4
votes
1answer
50 views

Why is the mean used to compute the expectation in the GAN loss?

From Goodfellow et al. (2014), we have the adversarial loss: $$ \min_G \, \max_D V (D, G) = \mathbb{E}_{x∼p_{data}(x)} \, [\log \, D(x)] \\ \quad\quad\quad\quad\quad\quad\quad + \, \mathbb{E}_{z∼p_z(...
0
votes
0answers
15 views

Are there any deep learning tools so that can be used to estimate long-tail distributions?

Are there any deep learning tools so that can be used to estimate long-tail distributions? update: my context is there are is there is a long-tail distribution (probability distribution) in my data of ...
1
vote
0answers
60 views

In continuous action spaces, how is the standard deviation, associated with Gaussian distribution from which actions are sampled, represented?

I have a question about implementing policy gradient methods for problems with continuous action spaces. Assume that actions are sampled from a diagonal Gaussian distribution with mean vector $\mu$ ...
1
vote
1answer
45 views

Intuitively, why can the training of a neural network be formulated as a probability estimation problem?

Neural network training problems are oftentimes formulated as probability estimation problems (such as autoregressive models). How does one intuitively understand this idea?
1
vote
0answers
41 views

Is the generator distribution in GAN's continuous or discrete?

I have some trouble with the probability densities described in the original paper. My question is based on Goodfellow's paper and tutorial, respectively: Generative Adversarial Networks and NIPS ...
3
votes
1answer
86 views

How does $\mathbb{E}$ suddenly change to $\mathbb{E}_{\pi'}$ in this equation?

In Sutton-Barto's book on page 63 (81 of the pdf): $$\mathbb{E}[R_{t+1} + \gamma v_\pi(S_{t+1}) \mid S_t=s,A_t=\pi'(s)] = \mathbb{E}_{\pi'}[R_{t+1} + \gamma v_\pi(S_{t+1}) \mid S_{t} = s]$$ How does $...
2
votes
0answers
28 views

What is the most efficient data type to store probabilities?

In ML we often have to store a huge amount of values ranging from 0 to 1, mostly being probabilities. The most common data structure to do so seems to be a floating point? Indeed, the range of ...
1
vote
1answer
81 views

Can a variational auto-encoder learn images composed of random noise at each pixel (each drawn from the same distribution)?

Can a variational auto-encoder (VAE) learn images whose pixels have been generated from a Gaussian distribution (e.g. $N(0, 1)$), i.e. each pixel is a sample from $N(0, 1)$? My gut feeling says no, ...
2
votes
1answer
66 views

What is the difference between model and data distributions?

Is there any difference between the model distribution and data distribution, or are they the same?
3
votes
1answer
59 views

What does a joint probability density function have to do with Stochastic Optimal Control and Reinforcement Learning?

I stumbled upon a job offer from a company that was looking for someone who was good with Reinforcement Learning (applied to finance) and something in their offer caught my eye. It goes something like ...
1
vote
1answer
72 views

How can I use the success and failure data to estimate parameters of a Dirichlet distribution?

I have used Beta function to estimate the performance of the agent. I have failure and success data of the task that runs on the agent. The parameter $\alpha$ is a number of successful tasks, while $\...
1
vote
1answer
254 views

Why do we regularize the variational autoencoder with a normal distribution?

When we define the loss function of a variational autoencoder (VAE), we add the Kullback-Leibler divergence between the sample taken according to a normal distribution of parameters: $$ N(\mu,\sigma)...
1
vote
2answers
112 views

Generalization error: Inputs drawn from distributions

I am currently studying Deep Learning by Goodfellow, Bengio, and Courville. In chapter 5.2 Capacity, Overfitting and Underfitting, the authors say the following: Typically, when training a machine ...
1
vote
1answer
56 views

Solving the supervised learning problem of learning $p(y \vert \mathbf{x})$ by using traditional unsupervised technologies to learn $p(\mathbf{x}, y)$

I am currently studying Deep Learning by Goodfellow, Bengio, and Courville. In chapter 5.1.2 The Performance Measure, $P$, the authors say the following: Unsupervised learning and supervised learning ...
4
votes
1answer
111 views

Many of the best probabilistic models represent probability distributions only implicitly

I am currently studying Deep Learning by Goodfellow, Bengio, and Courville. In chapter 5.1.2 The Performance Measure, P, the authors say the following: The choice of performance measure may seem ...
1
vote
2answers
90 views

Why does the machine learning algorithm need to learn a set of functions in the case of missing data?

I am currently studying the textbook Deep Learning by Goodfellow, Bengio, and Courville. Chapter 5.1 Learning Algorithms says the following: Classification with missing inputs: Classification ...
2
votes
0answers
73 views

Formulation of a Markov Decision Process Problem

Given a list of $N$ questions. If question $i$ is answered correctly (given probability $p_i$), we receive reward $R_i$; if not the quiz terminates. Find the optimal order of questions to maximize ...
1
vote
0answers
25 views

Why we multiply probabilities with support to obtain Q-values in Distributional C51 algorithm?

In 'Deep Reinforcement Learning Hands-On' book and chapter about Distributional C51 algorithm I'm reading, that to obtain Q-values from the distribution I need to calculate the weighted sum of the ...
1
vote
1answer
56 views

Why am I getting the logarithm of the probability bigger than zero when using Neural Spline Flows?

I am using a normalizing flow (Neural Spline Flows) to approximate a probability. After some training, the average loss is around 0.5 (so the logarithm of the probability = -0.5). However, when I am ...
1
vote
0answers
23 views

Deciding std. deviation for policy network output?

When I try to fit a Normal Distribution to the output of a policy network, for a continuous action space problem, what should be its standard deviation? mean for the distribution will directly be the ...
3
votes
2answers
607 views

What is a probability distribution in machine learning?

If we were learning or working in the machine learning field, then we frequently come across the term "probability distribution". I know what probability, conditional probability, and ...
1
vote
1answer
72 views

How does maximum approximation of the posterior choose a distribution?

I was learning about the maximum a posteriori probability (MAP) estimation for machine learning and I found a nice short video that essentially explained it as finding a distribution and tweaking the ...
5
votes
1answer
290 views

Why is Jensen-Shannon divergence preferred over Kullback-Leibler divergence in measuring the performance of a generative network?

I have read articles on how Jensen-Shannon divergence is preferred over Kullback-Leibler in measuring how good a distribution mapping is learned in a generative network because of the fact that JS-...
4
votes
1answer
77 views

In deep learning, do we learn a continuous distribution based on the training dataset?

At least at some level, maybe not end-to-end always, but deep learning always learns a function, essentially a mapping from a domain to a range. The domain and range, at least in most cases, would be ...
1
vote
0answers
19 views

Probabilistic classification - normalize results

I have a probabilistic classifier that produces a distribution over my 3 classes - C1, C2, C3. I want to compare some new points I'm classifying to each other, to see which one is the best fit for a ...
1
vote
1answer
81 views

Why is the entire area of a join probability distribution considered when it comes to calculating misclassification?

In the image given below, I do not understand a few things 1) Why is an entire area colored to signify misclassification? For the given decision boundary, only the points between $x_0$ and the ...
2
votes
1answer
189 views

What to do when PDFs are not Gaussian/Normal in Naive Bayes Classifier

While analyzing the data for a given problem set, I came across a few distributions which are not Gaussian in nature. They are not even uniform or Gamma distributions(so that I can write a function, ...
2
votes
1answer
47 views

Why is the expectation calculated over finite number of points drawn from a probability distribution?

This is from the book Pattern Recognition by Bishop. Why is expectation here a simple average? Why is $f(x)$ not being multiplied by $p(x)$?
2
votes
0answers
55 views

Which loss functions for transforming a density function to another density function?

I am looking at a problem which can be distilled as follows: I have a phenomenon which can be modeled as a probability density function which is "messy" in that it sums to unity over its support but ...
3
votes
1answer
82 views

How are the parameters of the Bernoulli distribution learned?

In the paper Deconstructing Lottery Tickets: Zeros, Signs, and the Supermask, they learn a mask for the network by setting up the mask parameters as $M_i = Bern(\sigma(v_i))$. Where $M$ is the ...
1
vote
0answers
65 views

Unit integral condition on the output layer

I want to train a neural network on some input data from a probability distribution (say a Gaussian). The loss function would normally be $-\sum\log(f(x_i))$, where the sum is over the whole data (or ...
2
votes
1answer
279 views

Why can we approximate the joint probability distribution using the output vector of an LSTM?

In the paper, Contextual String Embeddings for Sequence Labeling, the authors state that \begin{equation} P(x_{0:T}) = \prod_{t=0}^T P(x_t|x_{0:t-1}) \end{equation} They also state that, in the LSTM ...
3
votes
1answer
87 views

Standard deviation of the total input to a neuron

Raul Rojas' Neural Networks A Systematic Introduction, section 8.2.1 calculates the standard deviation of the output of a hidden neuron. From: $$ \sigma^2 = \sum^n_{i=0}E[w_i^2]E[x_i^2] $$ When I ...
1
vote
1answer
75 views

Binary vector expected value

Raul Rojas' Neural Networks A Systematic Introduction, section 8.2.1 calculates the variance of the output of a hidden neuron. Raul Rojas says that "for binary vectors we have $E[x_i^2] = \frac{1}{3}$...
7
votes
1answer
574 views

What loss function to use when labels are probabilities?

What loss function is most appropriate when training a model with target values that are probabilities? For example, I have a 3-output model. I want to train it with a feature vector $x=[x_1, x_2, \...
2
votes
1answer
56 views

Can the normalization factor for the belief state update be zero?

In order to update the belief state in a POMDP, the following formula is used: $$b'(s')=\frac{O(a, s', z) \sum_{s\in S} b(s)T(s, a, s')}{\mathbb{P}(z \mid b, a)}$$ where $s$ is a specific state in ...
1
vote
2answers
61 views

Can we derive the distribution of a random variable based on a dependent random variable's distribution?

In the diagram below, there are three variables: X3 is a function of (depends on) X1 and X2, ...
2
votes
0answers
54 views

What kind of distributions can be used to model discrete latent variables?

If we take the vanilla variational auto-encoder (VAE), we $p(z)$ is a Gaussian distribution with zero mean and unit variance and we approximate $p(z|x) \approx q(z|x)$ to be a Gaussian distribution as ...