Methods and principles of building "computer systems that automatically improve with experience."

learn more… | top users | synonyms (1)

0
votes
0answers
2 views

What are good tutorial on Weighted Finite Automata?

I would especially appreciate papers, books or tutorials with source code already available. Currently I'm reading "Spectral Learning Techniques for Weighted Automata, Transducers, and Grammars" by ...
-1
votes
0answers
10 views

clustering 1d data with a competitive learning method [on hold]

I have a 1D data, and I want to cluster that with any data using competitive learning method. How can I manage this? Thanks,
0
votes
0answers
6 views

Strange spikes in cost function during full-batch gradient descent

I was wondering what can cause spikes in the cost function (during training) similar to this one in the plot below: I am using a full-batch gradient descent, so as far as I know, with proper ...
0
votes
1answer
27 views

Best ways to model 'big data' given limited computing resources

Suppose I have a large data set(10 GB), with a response variable and multiple independent variables. What is the best way to utilize the data to build a model on? If the full data set includes 10 ...
0
votes
2answers
36 views

statistical prediction [on hold]

I am new to statistical methods so this question may be very simple for you. I want to know some "statistical prediction methods" for a sequence of numbers. The numbers may represent financial or ...
1
vote
1answer
21 views

Prediction uncertainty intervals for predictions of machine learning algorithms

Assume I have a regression problem. I fit models on a train data set and tune their hyperparameters using CV. I then run the models on the test set. What is the best way to calculate prediction ...
1
vote
1answer
15 views

out-of-bag error estimate for Boosted Trees

In Random Forest, each tree is grown in parallel on a unique boostrap sample of the data. Because each boostrap sample is expected to contain about 63% of unique observations, this lefts roughly 37% ...
1
vote
0answers
24 views

Batch Gradient Descent - Parallelization

Here is the scenario that I have been thinking for sometime: You have a large dataset over which you want to run a regression analysis, using the classic gradient descent method to minimize the error ...
0
votes
0answers
13 views

how to improve linear regression model [on hold]

i am working on a simple linear regression model for practicing in order to learn machine learning . my model runs correctly however it get a bad score which means it is a bad model so any advice for ...
1
vote
1answer
38 views

Feature engineering

I recently realized, that feature engineering (designing input vectors for machine-learning algorithm) is one of the most complicated tasks when applying known algorithms (for example kernel ...
0
votes
1answer
29 views

Adding observations by aggregating existing data

Question I'm aware that generating features from existing data can be a valid method for adding new features for a regression/ML algorithm*, but can you add observations generated from existing data? ...
1
vote
0answers
27 views

Generalization error of PCA and kernel PCA

I've been recently reading Shawe-Taylor et al. 2005, On the Eigenspectrum of the Gram Matrix and the Generalization Error of Kernel PCA, where the authors analyze the squared residual of kernel ...
0
votes
0answers
16 views

How can I calculate the AUC for softmax classifier (e.g., logistic regression)?

At the end of a convolutional neural network(CNN) , there are usually a softmax classifier attached to it. How can I calculate the AUC for the CNN (that is, for the softmax classifier)? Thanks!
2
votes
2answers
59 views

Representative elements of a set

I'm looking for the technical name of the following problem. It sounds like a standard machine learning technique, but I'm not familiar with the field, and can't seem to find it. Let's say that we ...
0
votes
1answer
34 views

Neural Network Forward Propagation

I'm trying to solve this neural network problem found here: How do I go ahead and calculate the forward propogate in this example? I've see examples of how to calculate the expected output but ...
3
votes
1answer
45 views

How to to find and evaluate optimal discretization for continuous variable with $\chi^2$ criterion?

I have a data set with continuous variable and a binary target variable (0 and 1). I need to discretize the continuous variables (for logistic regression) with respect to the target variable and ...
0
votes
0answers
19 views

The effect on learning [on hold]

What is the effect of providing the learning system with more data of some kind than the other? Will it become more effective at recognizing the first one than the other? For example, consider any ...
0
votes
0answers
24 views

How to stack a convolutional autoencoder?

I am trying to figure out as to how to stack a convolutional autoencoder (CAE)? Consider a convolutional autoencoder (CAE) (using MNIST data, 28x28 input dimensions): ...
0
votes
0answers
7 views

Heuristics for streaming data matching [duplicate]

I have an index composed by thousands of documents. Slightly modified copies of those documents are sent to my application in small chunks, and I need to check, from those chunks, which document has ...
3
votes
0answers
24 views

Heuristics streaming data matching

I have an index composed by thousands of documents. Slightly modified copies of those documents are sent to my application in small chunks, and I need to check, from those chunks, which document has ...
0
votes
1answer
48 views

Which machine learning model is applicable to the following case

I want to build a model that recognizes the species based on multiple indicators. The problem is, neural networks (usually) receive vectors, and my indicators are not always easily expressed in ...
0
votes
0answers
5 views

Guarded Discrete Stochastic network to solve CSP

There is a lot about NPL and vision systems but not a lot about constraint satisfaction problems (CSP). I have been digging around but I could use some guidance on how to use neutral network to solve ...
5
votes
1answer
42 views

What if high validation accuracy but low test accuracy in research?

I have a specific question about validation in machine learning research. As we know, the machine learning regime asks researchers to train their models on the training data, choose from candidate ...
0
votes
1answer
38 views

What is pretraining and how do you pretrain a neural network?

I understand that pretraining is used to avoid some of the issues with conventional training. If I use backpropagation with, say an autoencoder, I know I'm going to run into time issues because ...
0
votes
0answers
18 views

How to learn data fusion functions?

TL;DR: I have multiple datasets that say different/similar things (properties) about a bunch of entities (people, places etc.). I want to combine these into a single dataset that's closest to the ...
1
vote
0answers
28 views

Huge overfitting with Random Forests and Boosted Trees?

In the following picture, the boxplots represent a performance metric (the closer to 1, the better) recorded for 50 runs of cross-validation, and the black filled circles are the training values of ...
0
votes
0answers
9 views

How to perform pattern identification using ML?

I have the following problem: An event, takes place at a determined day of the week, hour, and with a pre-defined format (movie, music concert, lecture (3 items). Based on exit polls we determine 3 ...
0
votes
0answers
21 views

Overfitting in the validation set

When running an algorithm for training a system it is common to consider a lot of models and using the validation set for selecting one of them. In my case I am running a mini-batch gradient descent ...
0
votes
0answers
21 views

What will be the simple interpretation for the coefficients for features obtained in any Machine learning models?

I am working with a data that consists of two classes. I have used scikit learn, to craete models using SVM, Randomforest etc.I used to r2_score and I sorted the scores for features I am having and I ...
0
votes
1answer
17 views

Extracting Standard Errors Caret Model

I have tuned a glm net model with caret using the train function. I am trying to extract the coefficients and standard errors of those coefficients for the best tuned model. Following this CV post I ...
4
votes
1answer
22 views

Will normalizing training and testing data separately cause under/overfitting?

Suppose I have training and testing data and I want to train a classifier (e.g. SVM). Typically, features are normalized before classification to ensure some features aren't weighted more heavily than ...
0
votes
0answers
12 views

Prove Reccurrent Neural Network can exhibit oscillatory behavior

I understand how recurrent neural networks work, however I'm trying to build a deep intuitive understanding of their behavior which is difficult for me because they exhibit such complex behaviors. ...
1
vote
0answers
12 views

Interpreting crfsuite output model for numerical features

I am using crfsuite-python to implement a linear chain CRF in which I would like to use numerical features rather than strings as is the case with the standard CRF application parts of speech tagging. ...
4
votes
0answers
53 views

Ideal statistical or machine learning technique to model highly cross-correlated data

I'm trying to build a model that can predict streamflow for an alpine (snowmelt-fed) watershed using snow albedo (roughly, the energy reflectance of the snow) data. Albedo controls the melt of the ...
2
votes
1answer
23 views

Need a little help understanding K-means++ seeding

I have been working on a project that involves using K-means clustering for generating adaptive palettes from images. I understand the general process of K-means clustering, and I understand the ...
2
votes
1answer
58 views

R: Finding relationships between 2 variables to determine any patterns in data

I am working on finding relationships/patterns between 2 variables (Type_A, Type_B). ...
-1
votes
1answer
29 views

Generative algorithms

If I understand slide 4 correctly, the idea here is that in order to compute $p(y|x)$ we can use the fact that $$p(y|x) = \frac{p(x|y)p(y)}{p(x)}.$$ Then $p(x|y)$ and $p(y)$ are calculated using our ...
5
votes
1answer
85 views

Do CART trees capture interactions among predictors?

This paper claims that in CART, because a binary split is performed on a single covariate at each step, all splits are orthogonal and therefore interactions among covariates are not considered. ...
3
votes
2answers
50 views

Are Random Forests and Boosting parametric or non-parametric?

From this excellent paper by Breiman, we can seize all the difference between traditional statistical models (e.g., linear regression) and machine learning algorithms (e.g., Bagging, Random Forests, ...
0
votes
0answers
9 views

Variable representations for faster learning convergence

My notes on machine learning state that transforming a classification problem from 2 classes, class A = 0, and class B = 1, to class A = $(1,0)$, and class B = $(0,1)$ leads to faster convergence in ...
1
vote
0answers
16 views

How could these two simple Bayesian algorithms be explained, simply? [closed]

count(this token in class) + 1 / count(all tokens in class) + count( all tokens ) and ...
2
votes
0answers
34 views

How to make use of less data of a particular class for better modeling?

I have a dataset, say 9000 rows, with some features. Around 8000 belong to class 1 and 1000 to class 0. So, if I am creating a model with any method say SVM, LR, Random forest the model has a tendency ...
0
votes
0answers
34 views

In machine learning, may I train correctly a neural network with input real data and output validation Boolean data?

I have a matrix made of ~ 100 rows and 12 columns. Each entry contains a real value. The first 6 columns refer to a particular concept (firstClass), the following 6 to another one (secondClass), and ...
1
vote
1answer
45 views
+50

What is the honesty condition for regression trees?

I have a question pertaining to Stefan Wager's "Asymptotic Theory for Random Forests": http://arxiv.org/pdf/1405.0352v1.pdf Wager first states that trees are "fully grown in the sense given training ...
0
votes
0answers
11 views

RBM hidden units becoming correlated

I am trying to train an RBM with 8 hidden binary units and 40 visible ReLUs. At first, I had issues with binary units becoming stuck due to the weight saturating, but I got rid of that problem by ...
-1
votes
1answer
24 views

What are the benefits for semi-supervised learning over unsupervised clustering? Or any limitations?

I have another question about semi-supervised learning vs unsupervised clustering, what are the benefits and limitations? I have got some data with labels and some without labels. I performed ...
1
vote
0answers
16 views

Semi-supervised learning vs supervised learning, what are the benefits and limitations?

Just wondering if any previous work compared semi-supervised learning vs supervised learning? Currently, I have got both datasets with and without labeling. And therefore, it is intuitive for me to ...
0
votes
1answer
19 views

Determine confidence in a CART model with factor (2 levels) response variable (using rpart)

I use the package rpartto model a classification/regression tree. I have the variables $x,y,s$ where $x$ is in $\{-1,1\}$, y is continuous in $[0,1]$ and $s$ is a ...
2
votes
0answers
27 views

Reinforcement Learning in Industry [closed]

This is my first post here I would like to start with a rather general topic of discussion. I have studied Reinforcement Learning during the university years and although I find it rather fascinating ...
1
vote
1answer
25 views

Expectation of squared error

In machine learning, we let $X$ be a real-valued input vector and $Y$ be a real number output, with joint distribution $P(X,Y)$. We are looking for a function $f(X)$ for predicting $Y$ given the ...