Query Selector–Efficient transformer with sparse attention

We present Query Selector - a sparse attention Transformer algorithm especially efficient for long-term time series forecasting. Time forecasting is an old and important area of statistical research with vast practical applications to solving real life problems. In recent years, there has been growing interest in applying Deep Learning algorithms modeling. However, the best performing algorithm, Transformer, shows some problems modeling long term due inherent weakness this solution i.e. self-attention mechanism. Memory requirements canonical architecture are quadratically sequence length. attempt address problem.