To understand how to set the value of alpha, we’ll look closely at the update equation. Note that alpha must be set to a number between zero and one. When alpha is set to one, the new estimate is just the most recent return where we completely ignore and replace the value in the queue table. If we were to set alpha to zero, we would completely ignore the return and keep the old estimate unchanged. This is not something that we’d ever want to do because then our agent would never learn. But it will prove useful to set alpha to a small number that’s close to zero. In general the smaller alpha is, the less we trust the most recent return when performing an update and the more we rely on our existing estimate in the queue table. You’ll soon get a chance to experiment with setting this value of alpha yourself.