Tell me about the linear regression algorithm: how is it structured, what is trained, and is there an analytical solution?
Machine Learning / AI
Implement an algorithm for connecting sequences, such as hash join?
Tell us about Batch Normalization: what it is and why it is needed?
How to implement automatic rejection of calls when more than 10 calls are received within 15 minutes from a user?
How to implement a counter with a key lifetime in Redis for an auto-rejection call task?
What is the purpose of a pooling layer?
Why are you leaving your current job?
What happens if you apply bagging to linear algorithms?
How to organize data auto-collection and training using Airflow?
What is bagging?
Where are you currently located geographically and from where will you be working?
How do you compare request and product embeddings to find relevant candidates?
Where and what needs to be stored to ensure fault tolerance in auto-call rejection tasks?
Are there any other restrictions on trees in Random Forest?
What type of JOIN is used by default when writing a simple JOIN?
Tell us about the project you worked on at your last job, what you did, and what was your part?
What models do we take as the base in bagging? What is the depth of the trees?
Tell about Word2Vec: what training tasks are used in it (CBOW, Skip-gram)?
What does the architecture of Word2Vec training look like? How does the model predict the missing word based on context?
Tell us about the architecture of BERT and how it was trained.