Explainable agents for less Bias in Human-Agent Decision Making