179 questions
No questions match those filters.
← All questions
Training & Fine-tuningHard· Gumbel· Reparameterization· Sampling
Why use Gumbel noise specifically for retriever trainin...
This is one of the questions in the full AI/ML interview bank. Pro unlocks all 1789 questions; Premium includes the same bank plus the highest daily Practice limit.
See plansThe max-stability property of Gumbel noise allows the perturbation to act as a reparameterization rather than a heuristic smoothing. While Gaussian noise is intuitive, it results in the probit model, which is mathematically intractable for multi-class choice probabilities. Bias in this setup comes from the temperature parameter and straight-through substitution, not the noise itself.