Questions about computer algorithms that automatically discover patterns in data and make good decisions based on them.

learn more… | top users | synonyms

0
votes
0answers
1 view

What is the need of re-sampling the image for HOG features?

I read Dalal and Triggs paper for HOG description and a blog Chris McCormick HOG regarding the same. The blog says that the image needs to be re-sampled at different scales to recognize different ...
3
votes
0answers
17 views

ML with time-series data. Am I doing this right?

I work in renewable energy. My company gathers a lot of data from equipment. This typically includes process data (such as transformer temperature, line voltages, currents, etc.) and discrete alarms ...
0
votes
0answers
8 views

Does marginalizing on a Bayesian network preserve its original independence assumptions?

I know that marginalizing over a Bayesian network causes changes to the graph (e.g. marginalizing node c in the V-structure given by $a \rightarrow c \leftarrow b$ results in $a$ and $b$ being ...
0
votes
0answers
7 views

Machine learning algorithm for discovering trends [on hold]

I have been working on a trending system like Twitter's hashtag trends. This is my schema: Word_ID, Word, Ratio, Count, Day_Of_Count I have been counting the ...
2
votes
3answers
45 views

How to calculate IV, EV and optimal k for K-means?

Could someone explain how to calculate the following 3 evaluative properties: Intercluster Variability (IV) - How different are the data points within the same cluster Extracluster Variability (EV) ...
0
votes
0answers
17 views

How can I change the text style?

There are algorithms to change the forms of words, to define the subject of the text, as well as the definition of his style (literary, business, scientific, etc.). Do you have any ready ideas or ...
-3
votes
0answers
25 views

Help to implement epsilon greedy for q learning

I am currently looking for a method that implements an epsilon greedy action selector, but can't quite understand how it works, or why people have implemented it the way they have... As far I ...
1
vote
0answers
22 views

Variable elimination in Bayesian network

I'm trying to check if my understanding of variable elimination is correct. Assume the above Bayesian network is factorized as: $p(a,b,d,e,l,s,t,x) = ...
0
votes
0answers
12 views

How to calculate total variability matrix?

I'm writing paper about speaker recognition using artificial neural networks and currently I'm stuck with one thing. There is a Gaussian Mixture Model (GMM) that we can use to represent speech and it ...
0
votes
0answers
21 views

How to design deep convolutional neural networks?

As I understand it, all CNNs are quite similar. They all have a convolutional layers followed by pooling and relu layers. Some have specialised layers like FlowNet and Segnet. My doubt is how should ...
0
votes
0answers
26 views

Supervised Machine Learning for Event Classification

I've got some general questions about supervised machine learning. I have many disparate systems, all of which generate event information. These could be things such as hardware failure, software ...
3
votes
0answers
48 views

Are coevolutionary “Free Lunches” really free lunches?

In their paper "Coevolutionary Free Lunches" David Wolpert and William Macready describe a set of exceptions to the No Free Lunch theorems they proved in an earlier paper. The exceptions involve ...
1
vote
0answers
24 views

Inductive Bias - Decision Tree Pruning as a Bias

I am trying to understand inductive bias and have been looking around to try and work it out. I found this which explains what it is briefly but I have struggled to find examples I understand which ...
1
vote
1answer
18 views

Difference between SNN RL and DNN RL?

In Reinfrocement Learning (RL) in Neural Networks (NNs), I've seen two approaches to Q-learning. The first is to tile the state space with basis functions using Spiking Neural Networks (SNN) to ...
4
votes
0answers
20 views

Are the Confabulation Theories of Thaler and Hecht-Nielsen Isomorphic?

Both S. L. Thaler and R. Hecht-Nielsen have set forth neural-based theories of "confabulation" applicable to machine learning. The essential mathematics of Hecht-Nielsen is set forth in his paper ...
0
votes
1answer
16 views

reinforcement learning in gridworld with subgoals

Andrew Ng, Daishi Harada, Stuart Russell published a conference paper entitled "Policy Invariance Under Reward Transformations: Theory and Application to Reward Shaping" There is a specific example ...
2
votes
0answers
32 views

two ways of calculating the entropy in attribute selection (decision tree)

The definition of the entropy is $$H(Y) = -\sum p(y_j)\log_2 p(y_j)\,.$$ Now my text book says to compute the entropy for each attribute we consider the grouping of the data by that attribute now ...
-1
votes
0answers
15 views

necessary and sufficient criterion for the solution of the dual problem in SVM [on hold]

I'm studiyng the article "Support Vector Machine Solvers" of Bottou et al.[1] and i've problems to understand the reason why the optimality criterion expressed by the inequality (11) at page 8 is ...
2
votes
1answer
75 views

How do I stop “cheating” in reinforcement learning (MLP+Evo. Algorithm)?

I have a two hidden-layer MLP. I am trying to teach it classification of the sine function. For instance, if there is an [x,y] point above the sine function, the ANN should classify that point as a 1. ...
0
votes
1answer
19 views

What ML methods exist to categorize signal from noise? Red noise? Spatially correlated noise?

Let's say we are given measurements of some sort. In many cases, it is safe to assume that noise is white noise, serially uncorrelated, and zero mean with some finite variance. But in other cases, ...
1
vote
1answer
29 views

Find Correlations in Vectors of symbols

Given a set of vectors, lets say that each coordinate is populated from an alphabet (meaning set of symbols, numbers, etc) (particular or shared alphabets are indistinct). Is there any standard ...
2
votes
1answer
68 views

Why Isn't This Outlier Score/Reconstruction Error Not Squared?

I was looking through a paper called "AI2 : Training a big data machine to defend", and saw this (http://people.csail.mit.edu/kalyan/AI2_Paper.pdf) $score(X_{i}) = \sum_{j=1}^{p} (|X_{i} − ...
1
vote
0answers
16 views

How to mitigate the hierarchical error propagation in tree-structured classification

Suppose we have a multi-class classification problem, where the number of classes $K \geq 3$ We use a tree structure of multiple SVMs to divide and conquer the problem, with one example in the figure ...
2
votes
1answer
32 views

What is the Best and easiest way to create a Classifer for Sentiment Analysis [closed]

Sentiment analysis using Machine Learning is a hot topic. In the present situation when a person doesn't have a problem in having the training data set then which way should we create the classifier ...
0
votes
0answers
16 views

How do I choose the initial features vectors for a Stochastic Gradient Descent trained SVD++ algorithm?

I'm reading the SVD++ Netflix Recommender Systems paper because I want to be able to properly assess this approach to building a recommender system. How do I choose the initial values of $q_i$ and ...
1
vote
1answer
14 views

Using a combination of spatial and non-spatial inputs for convolutional neural networks

I'm working on training a game AI using deep reinforcement learning to achieve specific examples based on pixel input and some additional state information. Naturally, I'm using a convolutional ...
3
votes
1answer
34 views

Machine learning algorithm for predicting binary decisions on a large, underrepresented dataset

I would like create a classifier which works on a relatively large (about 30k samples) dataset with circa 20 attributes and a binary decision, however such, which contains relatively small amount of ...
2
votes
0answers
13 views

How does a recurrent connection in a neural network work?

I am reading a very interesting paper on genetic algorithms which define neural networks. I am familiar with how a feedforward neural network operates, but then I came across this: Where node #4 ...
2
votes
1answer
25 views

How is the environment designed for testing a reinforcement learning algorithm?

I'm working on a project, and have a candidate algorithm which I'd like to test. Before I go any further, I need to get the hang of how to code the "structure" of the environment in which my system is ...
0
votes
0answers
16 views

machine learning of infinite discrete point distribution

Is there any standard procedure (building feature vectors) that could be used to build machine learning models based on discrete and infinite point distributions on a hyperplane (practically 2D 3D), ...
-1
votes
1answer
51 views

Turing tests and humans

How are the questions framed in Turing tests? I mean what factors would one consider before framing questionnaire for the Turing Test.How the questions should be framed to make the test unbiased for a ...
1
vote
0answers
26 views

KDD Machine Learning using K-NN Algorithm Classification Problem

I'm trying to solve a classification problem from the KDD cup archive of 2004. Details can be found here: KDD 2004 Archive I'm only dong the particle physics part. The description of dataset is as ...
4
votes
0answers
38 views

Google DeepDream Elaborated

I've seen a few questions on this site about Deep Dream, however none of them seem to actually speak as to what DeepDream is doing, specifically. As far as I've gathered, they seem to have changed the ...
0
votes
1answer
21 views

What does it mean to have a continuous action space w.r.t. to reinforcement learning?

Last time I posted this question I got criticised for not being specific enough, hence this is my second attempt at trying to understand what it means to have a continuous action space. Please refer ...
3
votes
1answer
53 views

Standardizing Data for Neural Networks

Let's say we have a data set with following features [age, sex, country, city, annual income] [35, male, USA, New York, 73000]. I came across the article which ...
1
vote
1answer
37 views

Is this some kind of hashing?

Say I have $n$ vectors $\{ z_i \in \mathbb{R}^D\}_{i=1}^n$ (where $n$ is very large and hence I can't do any calculation which scales as $n$) and I want to create $n$ vectors $\{x_i \in \mathbb{R}^d ...
0
votes
0answers
16 views

Most impactful factor in the given set of values

My question is somewhat similar to this question Need an algorithm to find the input factors that are most affecting the output but the answers does not solve my problem. My question is: I am ...
1
vote
0answers
11 views

Is SRM necessary to prove that a countable union of agnostic PAC learnable classes is nonuniformly learnable?

The following I believe is a direct proof of this fact. If a learner is tasked to be $\epsilon$-competitive with a hypothesis $h \in \mathcal H_n$, where $\mathcal H_n$ is agnostic PAC learnable, it ...
0
votes
1answer
66 views

characteristic vectors for systems

The question is motivated from a physics problem: Let's first discuss the 1D infinitely long discrete system on a lattice, a system can look like: system 1: ...(ABAC)(ABAC)(ABAC)... this leads to ...
1
vote
0answers
33 views

Good language for introduction to self-modifying algorithms? [closed]

So I am trying to find a language with which i can write code to build/search through deductive reasoning 'nets', as well as self-modify it's search algorithms based on information learned from these ...
0
votes
0answers
26 views

Structural risk minimization erratum in Understanding Machine Learning Theorem 7.4?

Firstly define for a hypothesis class $\mathcal H_n$ with uniform convergence with sample complexity $m_{\mathcal H_n}^{UC}(\epsilon,\delta)$ the following function: $\epsilon_n(m,\delta) := ...
0
votes
0answers
31 views

Is there a non-linear version of ICA?

"Independent Component Analysis" is this : someone is sampling a random vector $s \in \mathbb{R}^d$ such that all its components $s_i$ are mutually independent and $\mathbb{E}[s_i^4] < 3$ and the ...
1
vote
1answer
49 views

Expectation Maximization Algorithm for simple naive Bayesian network

I am trying to understand the following network A has two children - B & C (aka common cause) All the variables are binary and can be either 0 or 1. In data values are missing only for some ...
3
votes
2answers
39 views

Which algorithm for counting the occurrences of a certain pattern (spots) in an image?

The Problem : Finding the number of occurrences of a certain pattern (or shape) in an image. In my example, the problem is about finding the number of spots (in variety of sizes) in an image. See ...
2
votes
2answers
53 views

How to handle missing continuous attribute values in ID3 (Iterative Dichotomiser 3)?

I'm implementing the ID3 algorithm (Iterative Dichotomiser 3). I have an attribute which happens to be continuous like 12.21, 3.01, etc. AND have missing values which are marked as "NA". How I'm ...
1
vote
1answer
47 views

Question about simple perceptron code

I'm reading through Sebastian Raschka's Python Machine Learning, and I see something confusing that is not explained in the text. In the code on this page: ...
3
votes
0answers
63 views

Typical NP-complete/hard problems in machine learning

I know little about machine Learning, but I work on optimization (solving NP-hard problems with SAT solvers or MIP). Examples of this would be solving TSP, Steiner tree problems, path finding with ...
0
votes
0answers
7 views

Modelling Congestion Control Problem as POMDP

I want to simulate a reinforcement learning based Congestion control algorithm. I saw http://lia.univ-avignon.fr/fileadmin/documents/Users/Intranet/chercheurs/habachi/TSP-2012.pdf I cant understand ...
2
votes
0answers
18 views

error measure (of ML agos) that takes confidence into account

when calculating the error measure such as mean absolute error, we use the real values and the predicted values. Many machine learning algorithms can give a confidence measure of each value being ...
1
vote
0answers
22 views

About the complexity of learning probabilistic graphical models

I guess that one way of measuring the complexity of learning a joint probability distribution is as its "sample complexity" (which is also sometimes known as its "distributional learning complexity"?) ...