Makintosz
  • Member for 2 years, 11 months
  • Last seen this week
  • Warsaw, Polska
Why do we need explainable AI?
4 votes

Another reason: In the future, AI might be used for tasks that are not possible to be understood by human beings, by understanding how given AI algorithm works on that problem we might understand the ...

View answer
If the accuracy of my current model is low ($50 \%$) and we want to minimize time in collecting more data, should we try other models?
2 votes

To know if your model needs more training data, try to plot out "learning curves", that are based on increasing size of the training set. Basically, you calculate training and validation accuracy ...

View answer
Price Movement Forecasting Issue
2 votes

Remember that any machine learning model works good only when there is a "rule" or a correlation between modeling data and modeled data. When there is not, even the best algorithm will not predict/...

View answer
Why can't we combine both training and validation data, given that both types of data are used for developing the model?
1 votes

There are two possible forms of overfitting. First related to the training only (fitting weights) and second related to architecture (fitting hyperparameters) and these two must be checked in two ...

View answer
What is the reasoning behind the number of filters in the convolution layer?
1 votes

From mathematical point of view you are correct as are your calculations. To catch all the patterns you need that many filters, but this is where a whole idea of a training comes in. Main objective of ...

View answer
Is randomness necessary for AI?
1 votes

It might be too philosophical answer, but maybe first we need to answer the question whether a human way of thinking or his creativeness includes random elements. For example if an author writing a ...

View answer
Accuracy too high too fast?
1 votes

It can be normal and there might be nothing wrong with your model. If there is a very strong and clear correlation in your data(good separability) then a network can achive very high accuracy very ...

View answer
What happens to the channels after the convolution layer?
Accepted answer
0 votes

Answer to my question is that values obtained from convolutions among different channels sum up together, therefore 3 channels after convolution with one filter give one output. Best explanation ...

View answer
Price difference predictions curve almost vanished
0 votes

I have encountered similar problem while trying to predict forex prices. I understand it this way: The data based on which you try to model price differences are so poor that the lowest error is ...

View answer