This repository contains code, data, and results accompanying our publication, Autoencoder-based feature extraction for the automatic detection of snow avalanches in seismic data. In this study, we applied unsupervised representation learning methods, concretely autoencoders, for the first time to seismic avalanche signals to automatically extract meaningful features. We benchmarked them against our baseline, a set of expert-derived seismic features, by evaluating the features on an avalanche classification task using random forest models. With this approach, we aim to advance and automate avalanche detection using seismic monitoring systems.