1 paper
Giovanni Compiani, Ilya Morozov, Stephan Seiler
We propose a demand estimation approach that leverages unstructured data to infer substitution patterns. Using pre-trained deep learning models, we extract embeddings from product…