Stack Exchange Network

Stack Exchange network consists of 175 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 [machine-learning]

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

1
vote
0answers
18 views

ANN webpage classification based on visual segmentation: sample size and parameterization

I am trying to classify web-pages based on their visual layout. My application visits a categorized webpage, catches certain elements' bounding boxes (rect in web-element term) and label them. Label ...
2
votes
1answer
43 views

Uncertainty and confidence in classification

I have a multi class classifier with 30-40 different classes. When i predict a class i would like to get an estimation about how certain my model is with its own prediction of that single sample . i....
0
votes
0answers
19 views

How to approximate continuous function from discrete uneven interval value?

To clarify, I have a discrete value of an unknown distribution. The value of this distribution domain is from 0 to 1. But the value which the discrete value is sampled has an unknown interval. For ...
1
vote
1answer
27 views

What are the theoretical/practical reasons to use normal distribution to initialize the weights in Neural Networks?

I'm aware that there are many different practices of initializing the weights when training a neural network. It seems traditionally standard normal distribution is the first choice. Most articles I ...
0
votes
0answers
35 views

Covariance Matrix of HIERARCHICAL MULTITASK GAUSSIAN PROCESS

I'm currently trying to develop a Gaussian Process to predict different levels of different individuals over time. So it is a Time Regression Problem in which we have multiple tasks, but also ...
1
vote
1answer
37 views

Between-subject fMRI classification: subjects with different number of runs

The main purpose of my work is to discriminate patients vs healthy controls using fMRI and multivariate pattern analysis (MVPA). Since I want to classify at the subject level I performed a separate ...
2
votes
1answer
36 views

Detection of music note sequence in audio signal

I have an audio signal which contains the combination of different western music notes(I know this combination in advance) and I want to identify the sequence of the music notes present in it. For ...
0
votes
0answers
11 views

Rademacher complexity for any hypothesis class

Problem Rademacher complexity is said to be the most powerfull measure of model capacity in statistical learning theory. But in statistical learning class we are just focusing on Rademacher ...
0
votes
0answers
15 views

LSTM models and Cross Validation

I have constructed a simplistic LSTM model using the keras package. The model takes sales data over a time series and tries to forecast future sales. The model I currently have groups the number of ...
4
votes
2answers
57 views

Train/test split that resembles original dataset and each other

I'm modelling a continuous variable (say, the average amount of smth per client). The variable has some asymmetric distribution: for example Gamma/Tweedie/ etc. Suppose that I'm not able to do cross ...
0
votes
0answers
17 views

Feature selection on training set without cross validation

I have a large dataset (1M+ samples) with 500 features. I need to create a predictive model that can be trained quickly. So, I want to perform an initial feature selection before building a classifier....
0
votes
0answers
17 views

HMM: Efficiently calculating a probability of a subsequence given the next hidden state

The forward-backward algorithm offers an efficient way to calculate the joint probability of a subsequence of observed states y1...yt and a hidden state xt = i (via alpha values). However, I am ...
-1
votes
0answers
18 views

does the normalized median vector of several percentage vectors represent the prototype/center

I have a few 3 dimensional vectors, which represent percentages and thus add up to 100: ...
0
votes
1answer
15 views

Grid search in r using caret package

I'm using the train function from the caret package to perform a grid search. (data is the spambase dataset from UCI Repository. I've added a row with headers) 1) Is this the way to perform a grid ...
0
votes
1answer
23 views

Can y of datasets be increasing/decreasing ratio when train regression model?

I found many cases in kaggle to predict the quantity or trends. They all set the real quantity as y. But I have question is that does anyone set the changing ratio as y? Like: ...
0
votes
0answers
10 views

Machine learning: eval_function for multiclass classifcation

I have checked What are the measure for accuracy of multilabel data? but I think the discussion somehow confused between loss and evaluation function. Say I have a multi-class classification problem ...
0
votes
0answers
17 views

How to methodologically show that a given 'time-series/sequential' data is not really sequential?

I have an apparent time-series/sequencial (supervised: multi-class classification problem) dataset with each data-point time-stamped. However, some domain intuition tells me that the data is static ...
1
vote
0answers
21 views

Dummy variable method for missing data in ML/predictive models

I'm looking for references on the use of zero-imputation with dummy-variable augmentation in the context of predictive models and MNAR missingness. Basically, the idea is that one imputes zero for ...
0
votes
2answers
22 views

Clustering for pair-wise data

I have been looking for algorithms in Python to create clusters from pairs of data. The data structure is similar to the stackoverflow post: An algorithm to create clusters from data pairs in python, ...
0
votes
0answers
14 views

What's different between hinge loss and squared hinge loss in SVC?

Sorry for the bad english. I'm an Asian. As the title says, What's different between hinge loss and squared hinge loss in SVC and how they effect on the decision boundary? And could you tell me when ...
1
vote
3answers
33 views

Why does SVM model need a test set?

First, sorry for the bad english. I'm an Asian so I'm not familiar with english. I'm studying SVC and I understood that a decision boundary of SVC only uses subset data of entire data set, which are ...
1
vote
0answers
25 views

Batch size stateful LSTM in Keras - how to evaluate with just one sample? [closed]

I'm using Keras to classify time series of 1000 timesteps containing one feature. In keras the input shape of an LSTM is defined as (batch_size, timesteps, features). So my input is f.e. (32,1000,1), ...
0
votes
2answers
39 views

What is a 'classification score' in machine learning?

I wanted to implement machine learning algorithms (SVM and KNN) for a biometric-based authentication system to classify whether a user is genuine or not. I have the feature data. What I understood, ...
0
votes
0answers
12 views

At which threshold is unbalanced data a problem for a binary classification tree? [duplicate]

I want to build a binary classification tree to clasfiy wether a person is working or not and use the model for prediction. I read that unbalanced data could be a problem. Now i ask myself at which ...
1
vote
1answer
50 views

ML Combination of Clustering and Regression [closed]

I recently faced a question that I, strangely, never thought of... Basically, I am considering a setup in which I want to perform regression on a very rich dataset. The chances to overfit are really ...
1
vote
0answers
23 views

Understanding probabilistic inference graphs

I am having trouble understanding inference graphs. In the diagram below I understand the graph on the left (forward graph) where the arrows describe the direction that data flows when training for ...
0
votes
1answer
20 views

Difference between ELMo and a normal deep network

I was reading up about ELMo and what I could gather was that we essentially combine the weights from different stacked lstm/gru layers for a given token as different layers are suited for different ...
0
votes
0answers
12 views

Statistical analysis of groundwater model outcome

I have simulated a groundwater model on an integrated numerical model software (Mike She) and calibrated it. My next step is to weigh the effect of each parameter used in the model by simply varying ...
0
votes
0answers
18 views

Using active learning to polish the accuracy after learning from a dataset

I am trying to create a prediction model which would return a numerical score upon receiving a set of representative numbers. The purpose was to replicate human quality assessment procedure upon ...
0
votes
0answers
15 views

Build customer ratings from subscription datetime data for a recommendation system

I want to build a recommendation system with only some customer's subscription and unsubscription date. I have a database that looks like: ...
1
vote
1answer
72 views

Machine learning & mathematics - [closed]

I am not from a statistics (or mathematics) major background. Just bachelors in computer science but had learnt algebra, probability, statistics, Integration & Differentiation, Permutation & ...
0
votes
0answers
23 views

Variation in LSTM Model based on seeds

I am working on a time series project. The results of LSTM model which i am using varies a lot with the variation in seeds. I am wondering how can i make that model stable. Currently to get the ...
-1
votes
2answers
34 views

What is the relation between the training time and the batch size?

In the deep learning process, especially SAR-ATR (e.g., generic object detection), is there any relation between the training time (speed) and the training batch size? Is there a paper related to ...
0
votes
0answers
9 views

Seperate Data by using hinge loss

It is more a theoretical question: It is given the labels or targets $t\in \{-1,1\}$ and the related prediction $\hat{y} \in [0, \delta_{max}]$. I want to find a threshold $\vartheta \in [0, \...
0
votes
1answer
13 views

How to reconstruct negative acceleration values using a simple autoencoder?

I am trying to reconstruct the acceleration values of a tri-axial accelerometer using a simple autoencoder. As acceleration values are often negative (e.g, -3.4) therefore using a ReLU activation ...
0
votes
0answers
16 views

What is a Rank in AUC in a classification method?

I have read a post on AUC for which the link is given below What does AUC stand for and what is it? I am putting a new post as the above is rather old. The person gave a few points related to Rank ...
0
votes
0answers
10 views

When do we use precision/recall over accuracy when evaluating a classifier? [duplicate]

When do we use precision/recall over accuracy when evaluating a classifier? What kind of scenarios would mean precision/recall metrics would be better than accuracy?
1
vote
0answers
23 views

Totally confused about input_shape dimension of stateful LSTM Network with keras [closed]

let's say I've got 2500 time series of class A, and 2500 time series of class B, both 1000 time steps long. There is only one numerical feature per time series. I want to train a LSTM using Keras to ...
1
vote
0answers
24 views

Gaussian Process regression: does there exist a conjugate prior over hyperparameters?

When adopting a fully Bayesian hierarchical setting in Gaussian Process regression is there a choice of kernel (covariance) function such that there exist a conjugate prior? If so which?
1
vote
1answer
31 views

Finding similar text - algorithms and evaluation

I've been asked to create a program that will rank similar texts to an input text given a collection of text. So far I've been using a tdidf representation and cosine similarity with a lot of regex-...
2
votes
1answer
23 views

deep learning : how to retrain an already trained model on different data?

I have a model (a convolutional neural network) that has already been trained on set of training data for human pose estimation. It works very well for simple cases, but I want to make it more robust ...
1
vote
0answers
11 views

Prediction problem in normal regression models and survival analysis cox PH [closed]

I am trying to predict the duration it takes for gas pipes to leak. I used 15 features which the most important one is “pipe installation year”. The latest leak data that I have is for a leak that ...
2
votes
2answers
480 views

Why uppercase for $X$ and lowercase for $y$?

Why is it most of the time (in many websites, articles or demonstration) the feature variable (columns) is denoted by a upper-case 'X' whereas the target variable is a lower-case 'y'? Looks more like ...
0
votes
2answers
49 views

Is there a way to check for sure that optimization was applied to the weights of NN?

Given the first or next steps of NN training, where weights were initially randomly sampled from some distribution, is there a reliable way to determine if some optimization was applied to the weights?...
0
votes
1answer
38 views

Using separate models to predict unbalanced classes

I'm facing a scenario with 5 classes where a tabulation of the target variable yields: > 1 2 3 4 5 > 1010 1310 1080 2700 2620 As you can ...
2
votes
1answer
99 views

How bad is Cholesky decomposition for OLS?

I've heard that I should never use Cholesky decomposition for OLS as it's super unstable, so I thought I'd try a very simple example to see how bad it could be. First let's set up the problem in R: <...
0
votes
0answers
37 views

do not need many controls with big data?

I am looking at a paper which uses a large panel data, 1 million observations, a dozen variables. I recall that in a discussion another one has the following comments: In structural models like ...
0
votes
0answers
15 views

Training on cifar100 [duplicate]

I am building a model on training a cifar100 images I try to follow others work on training cifar10 and they work very well(around 90% accuracy), so I add some more layers for training cifar100.But ...
0
votes
0answers
21 views

What causes lots of changes in loss and accuracy while training

I train a cnn model on medical images, with each epoch i print the loss and accuracy for train set but there is a lot of changes in loss and accuracy rate, the accuracy plays between 0.5 to 0.96 for a ...
0
votes
0answers
19 views

How to use the mean of a pre-trained CNN?

I have some doubts on how to use the mean of a pre-trained CNN. For example, if I use VGG-16 net, before the classification of an image I have to subtract the mean of the dataset (ImageNet in this ...