An extensive experimental evaluation is performed on 5 different real-world datasets that shows that the learned house embedding improves the performance of the evaluated regressors in almost all scenarios comparing to their results using raw features. [Link to the paper](https://www.semanticscholar.org/paper/19bfa2549f1323d56bee5bad7c2f7cf6795fd3d5)
Using Attention Networks to Learn Representations for House Price Prediction
—
by