**Daily update | 22 February, 2022**

# StackExchange

#### Sum of sample given a priori knowledge of its maximum

**Source:** stats**Views:** 146**Score:** 6**Tags:** distributions conditional-probability discrete-data sum

#### Why are residual connections needed in transformer architectures?

**Source:** stats**Views:** 226**Score:** 5**Tags:** neural-networks transformers attention residual-networks

#### Automated feature selection packages - Python

**Source:** datascience**Views:** 145**Score:** 3**Tags:** machine-learning deep-learning neural-network classification feature-selection

#### Good performance on both training set and validation set, but poor performance on the test set

**Source:** stats**Views:** 21**Score:** 2**Tags:** neural-networks python conv-neural-network keras

#### Optimizing logistic regression with a custom penalty using gradient descent

**Source:** stats**Views:** 11**Score:** 1**Tags:** regression logistic regularization loss-functions gradient-descent