An Attention Free Conditional Autoencoder For Anomaly Detection in Cryptocurrencies
This is an incremental improvement for anomaly detection in cryptocurrency time series.
The paper tackled the problem of detecting anomalies in noisy cryptocurrency time series by proposing an Attention Free Conditional Autoencoder (AF-CA), which improved explanatory power and anomaly detection compared to an LSTM Autoencoder.
It is difficult to identify anomalies in time series, especially when there is a lot of noise. Denoising techniques can remove the noise but this technique can cause a significant loss of information. To detect anomalies in the time series we have proposed an attention free conditional autoencoder (AF-CA). We started from the autoencoder conditional model on which we added an Attention-Free LSTM layer \cite{inzirillo2022attention} in order to make the anomaly detection capacity more reliable and to increase the power of anomaly detection. We compared the results of our Attention Free Conditional Autoencoder with those of an LSTM Autoencoder and clearly improved the explanatory power of the model and therefore the detection of anomaly in noisy time series.