What is label smoothing in image classification?
Machine Learning / AI
Why is the 80/10/10 masking used in BERT masking (mask, random, original)?
What are the methods of dataset debiasing (reweighting, resampling)?
What libraries do you use for solving typical machine learning tasks?
What programming language do you prefer to write code in?
Do you have any interesting proposals right now?
What is stream-based active learning?
Do you prefer a fixed work schedule from 9 to 18 Moscow time?
Why is the accuracy of a random forest higher than that of a single tree?
What was the most difficult task to implement and why?
How to calculate the perplexity of a language model? What are its properties?
What are the features of deploying models on edge (memory, latency, batch=1)?
What is meta-learning (MAML)?
What is visual servoing?
What are neural networks?
How did you get into the field of machine learning?
What data deduplication techniques are used in a pipeline?
What calibration methods for probability exist and how do they work?
Why did you decide to change your job?
What is a naive Bayes classifier?