Alexander Soare's user avatar
Alexander Soare's user avatar
Alexander Soare's user avatar
Alexander Soare
  • Member for 4 years, 1 month
  • Last seen this week
13 votes
3 answers
2k views

Why can't scikit-learn SVM solve two concentric circles?

6 votes
2 answers
140 views

What would be a good objective function for a computer vision model that predicts rotation?

5 votes
1 answer
129 views

Is "regularization keeps weights small" a misleading explanation?

4 votes
1 answer
782 views

Should Scikit-Learn CalibratedClassifierCV isotonic mode use bucketed rates instead of the actual targets?

4 votes
3 answers
638 views

What's the speed bottleneck in sklearn.svm.SVC.predict?

4 votes
2 answers
203 views

How to conceptualise statistical entropy

2 votes
1 answer
637 views

Is label smoothing equivalent to adding a KL divergence term or a cross entropy term?

2 votes
1 answer
45 views

Do Bernoulli bandits need a different treatment if the rewards are sparse?

1 vote
1 answer
271 views

What are some machine learning frameworks for supervised clustering?

1 vote
0 answers
32 views

For binary classification learning problems, how should I label instances where I'm only 60% sure?

1 vote
1 answer
2k views

What is meant by "number of support vectors" in the SVM implementation of scikit-learn

0 votes
0 answers
32 views

Would a 3D CNN require less training samples than a corresponding 2D CNN?

0 votes
0 answers
201 views

Under what circumstances is a fully connected layer similar to PCA?

0 votes
1 answer
59 views

Should the loss terms in diffusion model training be weighted by the diffusion schedule?

0 votes
0 answers
45 views

What's a sensible way to put a credible interval on samples in [0, 1]?