What is weight-only quantization and activation quantization?
Machine Learning / AI
What is Lovasz loss and where is it applied?
What is A/B testing? How to apply it in the context of model comparison?
What is dropout?
What is MDP (Markov Decision Process) and what are its components?
How do you feel about front-end tasks if they come your way?
Why should the shuffle parameter be set in a training dataloader when training for more than one epoch?
What is NTK-aware scaling RoPE for extending the context window?
How exactly does BatchNorm use the saved statistics in `eval()` mode and why is this important for result stability?
What are the algorithms for dimensionality reduction?
What is Mask2Former and its query-based approach?
What does the MLOps pipeline include from training to deployment?
How does banner regeneration increase conversions?
Tell us about the pronunciation assessment task: how to approach improving a model that detects errors in phonemes and sounds?
What is BERTScore and why is it better than BLEU for evaluation of generation?
How to organize containerization and deployment of models?
Have you previously undergone system design?
Why are you considering a new offer now?