What problems are there in detecting long-tail classes?
Machine Learning / AI
What is global average pooling and why does it replace a fully-connected layer?
Which metrics are more important for evaluating recommendation quality: precision or recall?
How to convert items into embeddings and recommend based on them?
What are the methods of dataset debiasing (reweighting, resampling)?
How to ensure novelty and diversity of recommendations?
How to solve the problem of frequent recommendation of popular products in retail?
What algorithms are used for collaborative filtering?
Tell us about your experience with the web frameworks Django and FastAPI?
What are foundation models in medicine (MedSAM, RadGraph)?
What are the pitfalls in causal inference in systems with feedback loops (recommendations)?
What features are there for training models on audio (sample rate, duration)?
How does QLoRA differ from LoRA?
What is sliding window attention?
What are the limitations of pgvector compared to specialized VDBs?
What is TGI (text-generation-inference) and its features?
What are the quality metrics for code-LLM (pass@k)?
Как учитывать временной промежуток между покупками для релевантности товаров?
How is A/B testing conducted?
What are EMA weights and where are they used?