Questions tagged [machine-learning]

Machine learning algorithms build a model of the training data. The term "machine learning" is vaguely defined; it includes what is also called statistical learning, reinforcement learning, unsupervised learning, etc. ALWAYS ADD A MORE SPECIFIC TAG.

Filter by
Sorted by
Tagged with
1
vote
0answers
5 views

How to reduce RMS error value in regression analysis & predictions - feature engineering, model selection

There's this dataset containing the metadata of Twitch's top 1,000 streamers of 2020. You can have the details here. I am currently participating in a challenge to predict the values for Followers ...
1
vote
1answer
20 views

Question on labels in machine-learning

Suppose that $X$ and $T$ are both random variables. Where $T$ is a label. I just want to ask is tha finding probabilities on the label $P(T=t)$ or conditional probabilities $P(X=x|T=t)$ meaningful or ...
1
vote
0answers
10 views

Parameter 'C' cannot be optimized for 'nu-svr'? mlr3 with kernlab

I am trying to optimize an SVR model within the mlr3 ecosystem with the kernlab package and I am getting the following error: The parameter 'C' can only be set if the following condition is met 'type &...
1
vote
1answer
24 views

Brain tumour detection using CNN

I have a fairly basic mathematical and implementational understanding of ML algorithms and CNNs, and I am trying to think of an approach for this task: https://www.kaggle.com/c/rsna-miccai-brain-tumor-...
2
votes
1answer
21 views

Transfer learning for regression problems

I have trained a regression model with 7 features for a given problem. Now, I have another regression problem (quite similar to the previous one) where I have only 6 samples in hand, but with 3 more ...
3
votes
2answers
69 views

How to choose the mean and std when using KFold cross validation?

With reference to this post on feature scaling, and many tutorials out there, it is mentioned how we should avoid data snooping by performing say, feature scaling on the train set, and then use the ...
0
votes
0answers
9 views

Neural network missing features in prediction data

Lets say I want to create a model that predicts an outcome of a certain match. I use features to train the network that are know before a match starts (such as individual performance of each player in ...
1
vote
0answers
5 views

Is there any relationship between confidence ratio in association rule and Bayesian rule?

I am currently studying Association Rule and somehow I thought about if there is any relationship between confidence ratio in Association Rule and Bayesian Rule. My knowledge in Bayesian Rule is not ...
0
votes
0answers
12 views

GBM Algorithm / Gaussian Loss function

I am working with the gbm() package in R, in which one is required to specify a distribution for the loss function. For regression problems, one is typically required to specify "Gaussian". ...
0
votes
0answers
8 views

Do I need to make one or more predictors covariance stationary?

I'm working with the Chicago Crime Data set. I want my model to use, among other variables, commodities prices to predict a particular type of crime. Since the data represent were collected daily, ...
0
votes
0answers
30 views

Is this K-Fold Cross-validation approach correct?

I've seen that Train-Validation-Test set technique is discussed and there is no consensus. Some people don't differ validation from test. When I was studying this technique and K-fold cross-validation,...
2
votes
0answers
29 views

Why is my 1D CNN so bad?

TLDR: My 1D CNN is doing a really bad job classifying graphs. Here's more context: Note: I've tried adopting the advice listed here and here, but my CNN hasn't stopped overfitting. I've already tried ...
0
votes
0answers
16 views

Is it possible to use xgboost to maximize a sum?

I am just learning about machine learning (no pun intended). I have used xgboost's objective function binary:logistic to find probabilities from labels of 0s and 1s. Is there some way to guide the ...
0
votes
0answers
7 views

Does Weight of Evidence Need to be Monotonic the Whole Time?

I've binned a few of my variables and have plotted them to take a look at the monotonic relationship. A good amount of them dip in the very beginning, then consistently rise until they get to the top. ...
1
vote
0answers
7 views

Reinforcement learning: is a softmax policy actor-critic expected to work on mountain car?

I am following David Silver's RL course and I'm struggling to apply the Actor Critic concept to the Mountain Car environment. I am using a softmax policy with linear function approximation. I am also ...
0
votes
0answers
10 views

Continuation of Training for Neural Networks

originally posted in SO Artificial Intelligence, advised to post here instead Background I am developing a GAN model (based off this paper), and am trying to learn how to feed subsequent time series ...
0
votes
0answers
42 views

What should i do with this kind of data? (price from the ads and actual price)

There are two subsamples in the dataset - on one the target is real(valid), and on the other it is approximate (I don't know how it differs yet, on one sample the real price of an apartment, and on ...
0
votes
2answers
24 views

How to compare the performance of ARIMA and LSTM for time series forecasting?

I am a beginner at the task of time series forecasting. I am facing some challenges in comparing LSTM and ARIMA for soma datasets. I would like to know if there are some general expectations regarding ...
0
votes
0answers
39 views

Which machine learning or AI algorithm is best for this scenario?

I have a program which takes 3 inputs (a, b, c) and produces 2 outputs (x, y). The program is deterministic; if I give the same inputs, I'm guaranteed to get the same outputs. My objective is find ...
0
votes
0answers
20 views

Measure prosodic similarity using deep learning

I have a dataset of 12,000 audio recordings of nonnative learners imitating the prosody of native speakers (300 samples for each native speaker utterance). All the nonnative learners' attempts were ...
0
votes
2answers
38 views

Adding Noise to continuous and categorical features?

Assume we have a dataset of 10 features, (combination of continuous and categorical features). I wish to add noise to each features separately, can i use the mean and SD of that particular feature to ...
0
votes
0answers
20 views

Why is the cost function for generator in GAN flipped?

I am going through the original GAN paper. In the paper cost for minmax game according to eq(1) is minG maxD V (D,G) = Ex∼pdata (x)[log D(x)] + Ez∼pz(z)[log(1 −D(G(z)))]. However the generator cost ...
1
vote
1answer
15 views

Unseen Classes in Decision Tree algorithms

I'm trying to train an XGBoost algorithm in a multiclass classification case, where the number of classes is very high (~6000). I recognize that this is quite difficult to do as it is, however, my ...
2
votes
2answers
109 views

Which predictive model to use for distributions like this

I have a target variable that has the following distribution. I have tried the typical regression models such as logistic regression, ridge regression, catboost regression etc. but I'm thinking that I ...
1
vote
0answers
28 views
+50

Are models using satellite image inputs well-posed?

I am using machine learning to create a land use regression model. My inputs are geographic coordinates. These I use to extract 80x80 meter satellite images or maps to feed the model. Lets take the ...
1
vote
1answer
28 views

Can Deep Recurrent Networks (Modelling of Sequence / Panel Data) handle feature vectors of different dimensions?

I am currently learning on sequence modeling with the coursera course from deeplearning.ai about RNN in general as well as the GRU and LSTM. However I am now one week in and still not sure, if these ...
2
votes
0answers
22 views

Supervised for Unsupervised problem

I am currently working on clustering using R. I have formed clusters using Hclust and PAM. Now, I am trying to find the feature importance for each cluster. As I have features and cluster information, ...
1
vote
0answers
13 views

Why Does using a OneVsRest Model for multiclass problem result in overall low accuracy but high accuracy for each individual class?

I have a multiclass problem i.e. 4 labels 0,1,2,3. I used a OneVsRest model wrapped around an xgboost model. What happens therefore is that i train a model 4x for each class. e.g.: ...
1
vote
0answers
17 views

Ways of better regression performance - lower RMSE loss value [duplicate]

There's this dataset of the top 1000 streamers on Twitch at 2020. I'm currently solving a challenge problem, to predict the amount of Followers gained based on the other features of each channel. The ...
1
vote
0answers
28 views

How do total rewards considered in reinforcement learning setting?

I am new to reinforcement learning and struggling to understand the basic concept of how the reward is calculated. Let's say I have 10 users. At each time step, different news articles are recommended ...
1
vote
0answers
15 views

Divide training data into multiple group, and train seperately

If I have a training data, I divide them onto many group. Then, for each group, I use a different neural network to train seperately/independently. Do you think this will be better than using a single ...
0
votes
0answers
9 views

Comparing two datasets with Average Precision/Precision Recall Curve

When comparing performances of classifiers between two different datasets, I use the average precision metric (the datasets are very imbalanced and thus ROC or just Precision unpreferable as was ...
2
votes
1answer
19 views

How to split training, testing and validation data for human subject experiment

I have experimental data of 25 human subjects. What is the efficient way to split it into train, test, and validation? It is given that,the number of data is different for different subjects. This ...
2
votes
1answer
57 views

Why do we need the concept of Risk in Bayesian Decision theory?

I'm studying Bayesian decision theory as introduction to machine learning and I see the concept of Risk in a lot of places. In the course I read, they define risk as: Risk is the expected error ...
0
votes
0answers
48 views

How to define relative error when the true value and its estimation are in [-𝜋, 𝜋)? [closed]

φ is the phase shift of a sinusoidal motion. φ^ is its estimation. φ, φ^ ∈ [-𝜋, 𝜋). The value of φ is set randomly. So, how to define an appropriate (relative) error Er to evaluate the estimation ...
0
votes
0answers
6 views

How can I download ImageNet dataset with only 20 or 30 classes? [closed]

I don't have powerful GPU to work on ImageNet dataset. I want to work on some classes of ImageNet in PyTorch.
0
votes
0answers
15 views

How to write a Proof-Of-Concept(POC) for machine learning model? [closed]

I've found that If any company is interested in your product, But they don't know it will fit, it will work or they don't trust you, They will ask you for a POC or Proof-Of-concept I need to write a ...
11
votes
4answers
1k views

Why is ROC insensitive to class distributions?

I am confused over why ROC is invariance under class distribution described in the paper An Introduction to ROC analysis. I cannot understand the example on why the proportion of positive to negative ...
0
votes
0answers
9 views

What ANN architectures can be used for variable length feature vectors (audio and video)? [closed]

How to input variable length feature vectors (audio and video) to the deep model, instead of fixed length input (e.g. x frames per second), for character recognition based on speech recognition based ...
0
votes
0answers
8 views

When data size is less than input size of a classifier [closed]

In my data, the input size of random forest is 2052 and the data size is only 300. In my case, the data size is less than the input size. But, random forest model makes good accuracy over 90 %. I ...
1
vote
0answers
233 views

Find the best threshold for logistic regression?

I am working on a customer purchase problem. I have 150 campaigns sent by email (or adds if you prefer), that I denote C0, C1 ......
3
votes
1answer
32 views

Operation modes in neural turing machine (Graves, 2014)

I am reading the paper "Neural Turing Machines" of Alex Graves (2014) and there are two points that are unclear to me. I would be very grateful if someone could help me out. More ...
0
votes
0answers
12 views

Are there other reasons to do offline training than computational efficiency in machine learning?

I am reading about online versus offline algorithms, and I cannot think of any machine learning algorithm that is truly offline. I mean, yes, neural networks are, for computational reasons, trained ...
1
vote
0answers
11 views

How to circumvent the Error: vector memory exhausted (limit reached?) when saving large Tidymodels workflowsets tibble [closed]

I have a question related to the methodology of working with tibbles and tidymodels workflowsets. I have a large number of models fitted and tuned using several recipe pre-processing steps/feature ...
0
votes
0answers
25 views

Conditional probability in Bayesian Network [closed]

In the following Bayesian Network how to calculate the probability of person having Bronchitis given that he has Dyspnoea
3
votes
1answer
210 views

Is sampling with replacement better than sampling without replacement?

This question is more specific to machine learning. Is sampling with replacement good for random forests because it leaves some out of bag samples for testing or is it because it creates datasets/...
0
votes
0answers
10 views

How to use discrete wavelet transformation to decompose non-time series data

I want to decompose x from the ames housing data using discrete wavelet transform (dwt) and integrate it to a linear regression model lm() to predict the sale price Here is the code I used ...
11
votes
1answer
240 views

Why is multicollinearity different than correlation?

I know that someone will probably say that this question is repeated and I will get a negative vote, but I'm very convinced that it's not, or at least it wasn't properly answered. See, we have lots of ...
1
vote
1answer
35 views

Will MSE-based estimator generate symmetric residuals if the error has got symmetric support (not distribution)?

This question is more specific than :my old question Take follow regression model: $y=f(x)+e$ Where $e\sim D$ with a such symmetric support $A=(-a,a)$, not symmetric distribution. Now given a data set ...

1
2 3 4 5
351