How is a random forest structured and how does it train in a classification task?
Machine Learning / AI
How do you compare the quality of different embedding models? What is MTEB?
Можно ли считать новую модель с F1-score 0.81 лучше модели с 0.8 и сразу запускать в продакшен?
What is the asymptotic complexity of multiple passes through a list?
If the AUC of a new model is 0.82 compared to 0.80 for the production model — can we confidently say that the new one is better? How to check statistical significance?
Is it necessary to regularize the bias term (w0)? Why is w0 needed at all?
How to measure the quality of a binary classification model? What metrics do you know?
What is H(X) in the context of classification, for example entropy?
How to compare floating-point coordinates considering the peculiarities of float in different languages?
Is it possible to cheaply fix the problem of module usage in gradient descent?
How do Precision and Recall behave under class imbalance? Example: 95 ones and 5 zeros, the model always predicts 1.
What is an ROC curve?
What is a re-ranker in RAG? What models are used (cross-encoder, bge-reranker)?
What is a branching criterion in trees?
What are your hobbies outside of work?
How to describe the architecture of a random forest model with an added linear model for extrapolation?
What is the Unigram language model in tokenization (like in SentencePiece)?