The tag has no wiki summary.

learn more… | top users | synonyms

1
vote
0answers
18 views

Learn the bayes net structure with latent variables while testing (but observed while training)

I want to use Bayesian network for data which has 5 types of variables which are inter-dependent on each other. Out of that, 1 variable is observed only while training but it is unavailable during ...
0
votes
0answers
6 views

Number of parameters in multinomial logistic regression

In Chapter 10 (Directed graphical models) of Murphy's Machine Learning text, the author claims that multinomial logistic regression has $O(K^2 V^2)$ parameters, where $K$ is the number of discrete ...
0
votes
1answer
44 views

Graphical dependence in the DAG X->Z<-Y

In Barber's book pp. 40-41 he says that the belief network X->Z<-Y: is "graphically dependent" since: $$p(x,y|z) \propto p(z|x,y)p(x)p(y)$$ I don't understand why graphical dependence follows ...
3
votes
1answer
31 views

Examples of marginal independence, conditional dependence

I am interested in finding "real-world" examples of when variables might exhibit marginal independence but are conditionally dependent given some other variable. It seems to me that the converse ...
0
votes
0answers
112 views

What distribution is the expectation taken over in the total expected pay-off in reinforcement learning? Is it consistent with Bellman's Equation?

I was following the reinforcement learning lecture notes on CS229: http://cs229.stanford.edu/notes/cs229-notes12.pdf on page 3 they have the equation for the expectation of the total pay-off: $$ ...
0
votes
0answers
7 views

Flow of influence in a v-structure for Probabilistic Graphical Models

I'm not very sure I understand why an observed v-structure have different flow of influence behaviour for a directed and an undirected graph. What is the intuition behind the actual definition for ...
2
votes
0answers
86 views

Minimizing KL divergence from a given distribution, according to a graph

Given $n$ discrete random variables $X_1,...,X_n$, a distribution $p$ on $X=(X_1,...,X_d)$ and a DAG (Directed Acyclic Graph) $G$ on $\{1,...,d\}$, which is the distribution $q$ factorizing with $G$ ...
1
vote
0answers
13 views

Using Bayesian Graphical Models to reconstruct duplicated damaged data

I am a computer science student specialised in machine learning. Recently I fell in love with Probabilistic Graphical Models (and probabilistic programming) because of the flexibility to focus on ...
0
votes
1answer
39 views

Mean field variational inference

In Chris Bishop PRML book p.465 equation 10.6, the derivation doesn't explain why exactly the term $\int q_j ln(q_j) dz_j $ was generated, is not that term supposed to be multiplied by constant, did ...
0
votes
0answers
11 views

How to Mine Tree Structures?

To learn similarities/differences between different instances (that are in the form of tree), what are the suitable methods/approaches? I know kernel methods and particularly tree kernels, but would ...
3
votes
1answer
47 views

How do you find mathematical expressions for the posterior marginals i.e. $P(x_n|y_0, … , y_n)$ in an HMM?

My goal is to find closed form equations for posterior marginals $P(x_n|y_0, ... , y_n)$ in a general HMM. I was told that we can calculate it exactly via BP (belief propagation, thought not sure how ...
1
vote
0answers
26 views

Comparing Factorie and Figaro languages for Statistical Relational Learning

I am looking to implement statistical relational learning, preferably in a modern programming language, and came across Factorie and Figaro for Scala. But most resources online that compare these are ...
0
votes
1answer
25 views

Message passing (belief propagation) in practice - observed variables

In a graphical model with variables with continuous distriubtions, and some observed variables, how can I compute the messages to be passed? I know the messages but I don't know how to implement it? ...
0
votes
1answer
26 views

Probabilistic Logical Graphical models like Markov Logic networks etc

I can't quite get a grasp of how and where these Probabilistic Logical Graphical Models (or PLGM or Statistical Relational Learning Models) score better than ordinary Probabilistic Graphical ...
1
vote
1answer
45 views

Does a Bayesian network include the CPTs?

I'm preparing slides for a lecture, and I require some guidance. I'm only talking about discrete variables. How would you formally define the concepts surrounding Bayesian networks? A Bayesian ...
0
votes
0answers
20 views

Best way to graph probabilities of feature vectors

I have a lot of feature vectors in the form of: v1=[x0, x1, x2, x3, x4] where x0, x1, and x2 can take binary values. either 0 or 1 x3 and x4 can take values from 0 up to 9 I have a lot of vectors ...
0
votes
0answers
43 views

stochastic network optimization

I'd like to optimize the flow of materials through a network. There are vertices (i.e. physical locations) and edges (i.e. links between the physical locations). Inputs: locations transactional ...
0
votes
0answers
11 views

Why do chordal graphs not lose conditional independences when its transformed from undirected to directed to factor graphs and around?

When chordal graphs are used to model probability distributions, why is it that they do not lose conditional independences when its transformed from a undirected to a directed to a factor graph and ...
0
votes
0answers
14 views

Why are chordal graphs special for inference in the context of Probabilistic graphical model?

I was trying to make a list of the reasons of why chordal graphs are important or interesting in the context of inference and probabilistic graphical models. Some of the reasons I have so far are: ...
2
votes
0answers
28 views

Efficient algorithm to enumerate all member DAGs of a Markov equivalence class

I'm working on a research project involving Bayesian networks. BNs are directed acyclic graphs (DAGs) used to compactly represent joint distributions of variables. In many cases, multiple DAGs can ...
0
votes
0answers
20 views

When would Probabilistic Graphical Model be more useful compared to other commonly used models?

When would PGMs be better compared to other classification algos like DT, or LR? I see that it will be better if there are relationships / dependencies between the features. Are there any other ...
0
votes
1answer
25 views

Factor graph vs Factor graphical model

In inference we use the terms undirected graphical models and directed graphical models. Why do we say factor graph instead of factor graphical models?
2
votes
1answer
45 views

How to perform Cross-Validation for glasso to select lambda in R

I am using glasso for variable selection. To get the best possible value of lambda cross validation is recommended. However, I am not able to find how to perform cross validation for glasso in R. ...
0
votes
2answers
68 views

What is the name of this type of tree / graph / representation?

I have come across a tree like way to organize variables and I wonder if anybody knows the name. A colleague thinks it has a "Japanese sounding name" (sorry by no mean I wish to be derogatory). So ...
1
vote
1answer
44 views

An example of r.v.s such that their distribution has more (conditional) independencies than their directed graphical model

I was trying to form an example where I had 3 r.v.s such that the distribution describing them had more conditional independencies or independencies than the directed graphical model corresponding to ...
0
votes
1answer
52 views

How does explaining away cause problems for learning?

In one of his lectures Geoff Hinton explains that a big problem of sigmoid belief nets is the explaining away phenomenon. I didn't fully understand this. I see that the induced width of the graph ...
0
votes
0answers
90 views

How to interpret residual vs fitted values plot with clustered points

I am performing a multiple linear regression and I have a plot of the my first two explanatory variables vs the residuals and also a plot with the residuals vs the fitted values. I am not quite sure ...
0
votes
1answer
37 views

Question of unary term (data term) of the graph cut method

I am trying to apply graph cut method for my segmentation task. I found some example codes at Graph_Cut_Demo. Part of the codes are showing below ...
1
vote
1answer
48 views

In Probabilistic Graphical Models, what does it mean that r.v. X influences r.v Y?

I wanted to pin down what the intuitive phrase: r.v. X influences r.v. Y as precisely and as rigorously as I could, and wanted to check if my interpretation was correct and complete with the ...
0
votes
1answer
50 views

Why do we need undirected (Markov) graphical models?

I understand the modular nature of directed models, and that each node captures a conditional probability. But why do we need undirected models? As far as I can see they lack intuition in that the ...
1
vote
1answer
24 views

Gaussian MRF/Markov Network: the zero precision = no connection?

Gaussian MRF in Gaussian information form: edge potential: $exp(\frac{-1}{2} y_s\Lambda_{st} y_t)$ node potential: $exp(\frac{-1}{2} y_t\Lambda_{t} y_t+\eta_ty_t)$ Why: precision parameter ...
2
votes
1answer
57 views

What is the role/purpose of hidden variables in graphical models?

Is there a formal treatment of the role/power of latent/hidden variables in graphical models and other machine learning models (e.g., structural equation models)? For example, the Restricted Boltzman ...
0
votes
1answer
107 views

about the definition of bayesian network

In this PDF http://people.csail.mit.edu/yks/documents/classes/mlbook/pdf/chapter2.pdf page 5 says: Given a set of functions $f(x_i,pa(x_i))$ non-negative and sum to 1, we define a joint ...
1
vote
1answer
33 views

Hidden Markov Model with conditional observations

I am looking for a research paper that basically describes a hidden markov model that has multiple observations, and some observations that have conditional dependencies. For example, please consider ...
0
votes
1answer
44 views

simply demostration on conditioned probability

I don't find the answer of this simply problem: how can I algebrically demonstrate that, with 3 variable A,B,C $ \sum\limits_C P(B|C)P(C|A)=P(B|A)$ under condition that $P(A,B,C)=P(A|C)P(B|C)P(C)$ ...
1
vote
0answers
32 views

How are deterministic/logical nodes represented mathematically in directed graphical models?

Various software for performing inference on graphical models support logical(a.k.a deterministic) nodes. PyMc, Winbugs, Smile are the ones I'm aware of. Based on the different inference methods ...
3
votes
2answers
84 views

Learning to map vectors to vectors

Say we want to learn a function: $f(\mathbf{x} \in \mathbf{R}^p) \rightarrow \mathbf{y} \in \mathbf{R}^q$ where $\mathbf{x}$ and $\mathbf{y}$ are vectors representing time series. We have multiple ...
0
votes
1answer
43 views

identifying latent variables in this model

I have been trying to understand EM and I am having a hard time understanding what a latent variable is. In particular, I am having issues in identifying whether in a particular model that I am using, ...
0
votes
0answers
46 views

Log linear models advantages

What are the advantages of log linear representation in opposite of table representation? Is it simply computational issue ( avoid overflowing)? For example, in a markov network A-B we can represent ...
2
votes
1answer
72 views

What is the correct definition for completness in d-separation in directed graphical models?

I was reading Koller's book of probabilistic graphical models and in section 3.3.2 she discusses what properties should hold for d-separation as a method for determining independence. She tries to ...
2
votes
2answers
142 views

What does the notation $(\textbf{X} \perp \textbf{Y} , \textbf{W}\mid \textbf{Z})$ mean?

I was reading Koller's and Friedman's Probabilistic Graphical Models book and became confused about some of its notation because of a set of notes that either contradict it or express it differently. ...
1
vote
1answer
116 views

LASSO or other regularized regression with censored (missing) data

Here is my problem. I am looking at various time series curves. Let's call them total spend aggregated over all customers on various products versus time. At any given time, I want to predict the ...
2
votes
1answer
814 views

Smoother lines for ggplot2

This question probably has a simple solution, still the thing is I've written a code to plot mortality in 2 different groups and that is, death in obese patients vs not obese. Now their are 2 groups ...
0
votes
1answer
24 views

Graphical Model Equivalent of Matrix Pseudoinverse

The may sound like a strange question but I was wondering if a Pseudoinverse of a matrix could be found using SVD whether there was a graphical modelling equivalent that could be used to estimate the ...
1
vote
1answer
36 views

Markov blanket conditional distribution derivation

I am trying to derive the formula for the conditional distribution for a variable in a Bayesian network: $$p(x_j|x_{-j})=p(x_j|x_{pa(j)})\prod_{k\in ch(j)}p(x_k|x_{pa(k)})$$ I understand D-separation ...
0
votes
0answers
50 views

Assumptions implied by “pairwise marginal” parameterization of MRF

I'm trying to understand the assumptions of different parameterizations in a Markov network. In this case, I'm trying to understand the assumptions (and effects) that result from parameterizing ...
2
votes
1answer
49 views

What is max-sum / max-product variant of loopy BP computing?

In (Nowazin and Lampert, Structured Learning and Prediction in Computer Vision, p. 29.), they say that in the max-sum variant of loopy belief propagation, the "variable max-beliefs are no longer ...
1
vote
0answers
28 views

How to design an energy function in a factor graph model?

When designing a factor graph, the designer needs to specify the structure of the graph (encoding independence assumptions), and the form and parameterization of the energy functions for the ...
1
vote
0answers
18 views

Does anyone know how to make this infograph with R? [closed]

This picture was posted on a Brazilian newspaper called "Folha de São Paulo". The graph relates each candidate's name with their allied political parties
0
votes
0answers
34 views

Possible inferences from a graph pattern

So, I had a weighted dynamic graph having info about 10 consecutive timesteps ( basically 10 files ). Now, I had to mine out patterns in the weight and structure of the complete graph. I did that. The ...