Vladimir Belik's user avatar
Vladimir Belik's user avatar
Vladimir Belik's user avatar
Vladimir Belik
  • Member for 3 years, 8 months
  • Last seen more than a week ago
16 votes
5 answers
3k views

For neural networks, is mini-batching done purely because of memory constraints?

6 votes
2 answers
2k views

Why is GARCH offering no predictive value?

3 votes
1 answer
1k views

Why is my stacking/meta-learning not outperforming the best base model?

3 votes
0 answers
1k views

How many features/dimensions/variables can K-Nearest Neighbors handle with a finite amount of data?

1 vote
1 answer
759 views

What do the "in/out features" of a neural network mean?

5 votes
1 answer
427 views

Why custom evaluation/scoring metric is causing overfitting in (cross) validation?

0 votes
0 answers
407 views

How many variables can logistic regression handle?

0 votes
0 answers
333 views

XGBoost: is tuning max depth effectively the same as tuning min_child_weight, or vice versa?

2 votes
1 answer
309 views

Reinforcement Learning - what causes a plateau in the performance of the model during training?

5 votes
1 answer
304 views

Can RandomForest multiplicatively combine features?

3 votes
1 answer
209 views

Can I use oversampling to increase sample weights?

2 votes
1 answer
203 views

Changepoints in linear regression (NOT piecewise regression)

0 votes
0 answers
185 views

How to get around the weaknesses of stationarity tests like ADF test?

1 vote
0 answers
110 views

How to create/design a Hidden Markov Model?

1 vote
1 answer
55 views

ML model trained on 2 sets of features - both sets deliver good OOS accuracy on their own, but combined result is same/worse than alone. Why?

0 votes
0 answers
47 views

How do HHMM (Hierarchical Hidden Markov Models) work and where to learn more?

3 votes
1 answer
31 views

Neural networks - what does learning rate exactly mean and is it applied over batches or epochs?