Questions tagged [supervised-learning]

Supervised learning is the machine learning task of inferring a function from labeled training data. The training data consist of a set of training examples. In supervised learning, each example is a pair consisting of an input object (typically a vector) and a desired output value (also called the supervisory signal). A supervised learning algorithm analyzes the training data and produces an inferred function, which can be used for mapping new examples.

Filter by
Sorted by
Tagged with
0
votes
0answers
4 views

On quantifying the amount of information per example provided to the model in Supervised vs Self-supervised learning

I've seen Yann Lecun in his self-supervised learning talks talking about how traditional supervised learning (Classification setting) by attributing a class out of N classes to each example the ...
0
votes
0answers
2 views

Accuracy of supervised versus unsupervised learning

In most articles I have read it's implicitly assumed that unsupervised learning yields inferior results to supervised learning. In the rest, the assumption is quite explicitly stated. Intuitively, the ...
0
votes
0answers
13 views

Find optimal training dataset after concept drift

There are many strategies how to detect a concept drift or model drift, like when there was a major change in the underlying process so that the model becomes invalid. It can be an abrupt change or it ...
0
votes
0answers
8 views

When would AUC fail in comparing models? [closed]

It is possible that a classifier might have 1 threshold where there is highest possible true positive rate and least possible false positive rate (and lets say that is what the application requires), ...
0
votes
0answers
7 views

Multi Target Techniques where Dependent Variables are Correlated

I have browser data that contains over 100 independent variables to predict customer spend. Instead of predicting total spend over a given time, let's say we want to predict the monthly spend for each ...
1
vote
1answer
18 views

What is the difference between Transfer learning and Trained/Supervised machine learning?

I am trying to understand the difference between the supervised / labelled machine learning and the trasnfer learning. From my reading and understanding they are similar. Because in both cases we use ...
1
vote
1answer
24 views

Topic modeling for regression

Is there a way to influence the way topics are created with topic modelling in the sense that the topics also reflect their influence on the target variable of a machine learning problem? I have a ...
0
votes
0answers
5 views

Supervised Discretization based on multiple time series

I'm having multiple time series observations $X_{k1},..., X_{kt}$ with a single binary response $Y_k$ for each time series $X_{kj}$ for $j = 1, ..., t$ (Multivariate Time series). Now, I want to make ...
1
vote
1answer
67 views

Q-function in Q-Learning

I ran into solved old-exam question as follows: My notes tell me that option b is correct but I think option d is correct. is there any idea why (b) is correct?
2
votes
1answer
52 views

Information gain of the root node

Recently I saw this question and answer as attached in following image Anyone can add details how this solution achieved?
0
votes
1answer
30 views

How to handle errors in your target data?

For the sake of the question, assume I'm presented with a simple classification task (for simplicity, let's assume binary classification). We are given a feature matrix $X$ and a target vector $y$ of ...
8
votes
3answers
268 views

Why does a function being smoother make it more likely?

I am currently studying the textbook Gaussian Processes for Machine Learning by Carl Edward Rasmussen and Christopher K. I. Williams. Chapter 1 Introduction says the following: Given this training ...
0
votes
0answers
30 views

Overfitting during last model training stage

To select, tune, train an ML model I used the following stages: Split data in train / hold-out dataset Perform nested cross validation using the train data only (loop with several models with a grid ...
1
vote
0answers
18 views

Which are some formal approaches for predicting multiple binary time series?

I have 10000 roughly similar individuals. For each individual I've got a response (binary time series), 200 explanatory features (also time series), and 10 static features that represent ...
2
votes
3answers
51 views

Confusion about Understanding Supervised Learning as Bayesian Inference

I am going through a lecture that is explaining how supervised learning can be thought of from a Bayesian perspective, where we are trying to maximize log p(theta | data). I am confused as to what the ...
0
votes
0answers
21 views

Using Survival Analysis to Estimate Loan Prepayment Behavior/Speeds

I have a dataset which consists of monthly account level information for fixed rate loans that were originated on or after January 2013. The account level information includes the account id, ...
1
vote
2answers
112 views

Semi Supervised learning vs Supervised

I am trying to understand the mathematical properties of supervised learning and semi-supervised learning. Let us consider the case for the mean $\mu$. Then the supervised learning estimator can just ...
0
votes
1answer
32 views

How is stratified sampling better than sampling equally from all classes while crossvalidating?

I can see that stratified sampling helps in maintaining the same class distribution in the training set as in the original dataset. However, my understanding is that ideally, the model should be ...
0
votes
0answers
20 views

I have a classifier, now I want to identify best parameters

I have data which I have classified using many supervised classifiers (using matlabs classification learner). I am classifying a Pass/Failed test vector with 50 different variables\features. I have ...
0
votes
0answers
16 views

Confidence Interval as a feature in supervised learning?

Imagine a model that predicts the probability that a given online ad will be clicked by a given user. One of the features is the click-through rate (CTR) of the user (...
17
votes
4answers
741 views

Why does regularization wreck orthogonality of predictions and residuals in linear regression?

Following up on this question... In ordinary least squares, the predictions and residuals are orthogonal. $$\sum_{i=1}^n\hat{y}_i (y_i - \hat{y}_i) = 0$$ If we estimate the regression coefficients ...
2
votes
1answer
34 views

Proper Scoring Rule in Optical Character Recognition

Cross Validated likes to promote proper scoring rules in "classification" problems. That is, get accurate probability predictions. Then make the classifications, taking into account the cost ...
1
vote
0answers
17 views

Which machine learning model? For small dataset and very long feature vectors [closed]

I've got two questions about which model might be the best for the assumption described below: In the assumption, we are given various of compounds, wanting to determine the result of the reaction of ...
0
votes
0answers
32 views

Changing L2 regularization constant in logistic regression proportionally to the number of columns/rows in the dataset

I'm trying to use scikit LogisticRegression to solve a multiclass text classification problem with variying number of columns (unigrams) in the trainging datasets. From what I understood, L2 ...
0
votes
1answer
11 views

Multi-regression model validation

I'm a new-bee in the ML modelling and have created a multi-linear regression model. I have got the rmse score for the model as approximately 5. How am I suppose to interpret this?
0
votes
0answers
7 views

Is there a machine learning method for matching similar groupings?

I have a problem where I have rows/samples that are grouped together and each sample has a specific label (my data is genetic with genes being the samples and they are grouped together in the genome ...
0
votes
0answers
34 views

Square loss for “big data”

Let’s set up a supervised learning problem with $p$ predictors and $n$ observations. The response variable is univariate. The problem can be regression or classification, though I think a ...
0
votes
0answers
12 views

What's the best strategy to fill NAs for a predictor in supervised learning e.g. SVM?

What's the best strategy to fill NAs for a predictor in supervised learning e.g. SVM? I have monthly data for all other predictors since 1963 and for one predictor I have data since 1990 only. So I ...
0
votes
0answers
12 views

What approach should be used to model changing conversion rates?

Suppose I want to predict conversion rates of products in an eCommerce web site. Conversion rates can change over time due to changing market condition in addition to seasonality. I can build a ...
0
votes
0answers
38 views

How to determine equation of hyperplane for SVM?

Assume we have only two features in our training dataset that is already classified into class C1 and class C2. The transposes of the feature vectors are given below for each class: C1: [2 6], [1 1], [...
0
votes
0answers
24 views

Extract Keyword/Concept From Column Description Using NLP

Suppose in my database, each table has a description associated with each column and I want to further extract keyword or key concept from the description. For example, mean of transaction amount in ...
1
vote
1answer
18 views

Would the softmax classifier ever yield equal probabilities for more than 1 class?

The question might be quite straightforward but I can't seem to be find any relevant resources from Google. All the sources I found are focused on explaining difference between softmax and sigmoid ...
2
votes
0answers
32 views

Best method to handle unknown class in supervised classification

I have a training dataset, where the records are labelled into 3 classes: A, B and C. My testing dataset consists of records that belong to classes A, B, C and records that do not belong to any of the ...
1
vote
0answers
13 views

General implications of low entropy for a dataset

I am fairly familiar with entropy, which quantifies uncertainty/surprisal of a random variable. In my case, I have a corpus where I can use empirical word frequencies to estimate entropy of the entire ...
3
votes
1answer
30 views

ROC Curve for unbounded scores

Say I have a classifier that assigns a score to an image based on whether it has a cat in it. The higher the score, the more likely there's a cat in it. But for this classifier, the value of the score ...
0
votes
1answer
22 views

RFE, feature selection for Churn/Credit Risk modelling

I am currently working on a redemption model for a financial company, using time series data and Logistic Regression. Currently we have a few features that are time dependant (I know, logistic is not ...
0
votes
1answer
30 views

Given a linear regression model with p predictors why is p the lower bound on the number of sample observations we need?

Working through notes that say the following: Given the linear model: $$Y=\beta_0+\beta_1X1+...+\beta_pXp$$ we need to collect at least n=p sample observations to infer the values of the parameters. I ...
0
votes
2answers
56 views

Can PCA be used for independent variables?

Currently I've been carrying out a research on prediction of some thermal properties of steels from chemical composition only. For that purpose I have 13 inputs (for instance: C,Mn,Ni, etc). They are ...
0
votes
0answers
12 views

What is the relation between Minimum description length, Model evidence and Shannon's source coding theorem?

Given samples $(x_i,y_i)$ drawn independently from $P(x,y)$, we usually have in supervised learning framework the objective of minimizing\maximizing an approximation : $$min_{w} \{ \sum_iL(y_i,f_{w}(...
1
vote
1answer
17 views

Supervised learning dataset for click prediction in real time ad bidding

I have a dataset of real time bidding events (programmatic advertising) with the following properties: Dataset spans a period of several consecutive weeks. 250M total bids. The winning percentage is ...
1
vote
0answers
19 views

KNN and Classification Factor

I'm studying the KNN algorithm and its application on R. If I correctly understood, KNN is a supervised algorithm able to classify an unlabeled item according to the predominant belonging class in the ...
0
votes
2answers
45 views

Relationship between structural or statistical properties and hardness of classification

I am trying to understand the relationship between structural or statistical properties of training dataset and hardness of classification in the context of binary classification with SVM using RBF ...
0
votes
0answers
59 views

How to prepare data for Bert fine-tuning?

I am looking at Bert documentation in order to fine tune a pretrained model... So lets say I have a new dataset for paragraph classification. I get a vector for each paragraph and I use a simple KNN ...
2
votes
1answer
32 views

Evaluating multiclass imbalanced problem per class

For a multiclass imbalanced problem, accuracy is not a good metric to evaluate model performance. Equally, accuracy is a global ...
3
votes
0answers
88 views

Graphical representation of Bayes decision boundary

Here is my problem statement: Let $X=(X_1,X_2)∈[0,1]×[0,1]$ and $Y∼Bernoulli(p=X_1⋅X_2)$. Plot the Bayes decision boundary ${(x1_,x_2):P(Y=1|X=(x_1,x_2))=0.5}$ and indicate the regions in $[0,1]×[0,1]$...
0
votes
0answers
18 views

Is it an overfitting problem for SVM classification?

I am new in Machine learning, and I want to detect emotions from the face. Preprocessing: I used equalizeHist to equalizes the histogram of grayscale images (JAFFE database with 213 images), in the ...
-2
votes
1answer
56 views

Extremely hard binary classification problem [closed]

Could anyone point me to a collection of binary classification dataset where a support vector machine algorithm will miserably fail?
0
votes
1answer
15 views

How to call the combined set of training, test and validation data

If I want to refer to the whole set of preprocessed data samples ready to be split into training-, validation- and test-set, how should I call it?
1
vote
1answer
29 views

How to deal with training models on data where the examples are highly dependent on each other?

Say you have a dataset of products sold at a store with the special condition that each day there is only one of each product in stock. That is, if there are multiple orders for a given product on a ...
0
votes
0answers
17 views

Latest research and explanation on how semi-supervised learning is performing better than supervised?

So in AAAI 2020 also semi-supervised learning is given the push. There are some intuitive reasoning provided by people but since the research is so fast, I wanted to know actually what is the latest ...

1
2 3 4 5
11