Interested to Buy Any Domain ? << Click Here >> for more details...
Answer Posted / Arvind Saini
The Leaky ReLU (Rectified Linear Unit) function is a variation of the standard ReLU function that aims to address the vanishing gradient problem in deep neural networks. Instead of outputting 0 for negative inputs as the standard ReLU does, the Leaky ReLU outputs a small positive value between 0 and some constant 'a'. The formula for Leaky ReLU is: f(x) = max(0, ax) + b. Here, 'a' is the slope of the line for negative values (usually set to 0.1), and 'b' is a bias term.
| Is This Answer Correct ? | 0 Yes | 0 No |
Post New Answer View All Answers
Please Help Members By Posting Answers For Below Questions
No New Questions to Answer in this Category !! You can
Post New Questions
Answer Questions in Different Category
