ID-Conditioned Auto-Encoder for Unsupervised Anomaly Detection
This is an incremental improvement for machine condition monitoring using sound data.
The paper tackles unsupervised anomaly detection by introducing an ID-Conditioned Auto-Encoder, adapting a class-conditioned approach for machine condition monitoring with sound data, and achieves results evaluated on ToyADMOS and MIMII datasets from the DCASE 2020 Challenge Task 2.
In this paper, we introduce ID-Conditioned Auto-Encoder for unsupervised anomaly detection. Our method is an adaptation of the Class-Conditioned Auto-Encoder (C2AE) designed for the open-set recognition. Assuming that non-anomalous samples constitute of distinct IDs, we apply Conditioned Auto-Encoder with labels provided by these IDs. Opposed to C2AE, our approach omits the classification subtask and reduces the learning process to the single run. We simplify the learning process further by fixing a constant vector as the target for non-matching labels. We apply our method in the context of sounds for machine condition monitoring. We evaluate our method on the ToyADMOS and MIMII datasets from the DCASE 2020 Challenge Task 2. We conduct an ablation study to indicate which steps of our method influences results the most.