Meta-supervision for Attention Using Counterfactual Estimation

Abstract Neural attention mechanism has been used as a form of explanation for model behavior. Users can either passively consume explanation or actively disagree with explanation and then supervise attention into more proper values (attention supervision). Though attention supervision was shown to be effective in some tasks, we find the existing attention supervision is biased, for which we propose to augment counterfactual observations to debias and contribute to accuracy gains. To this end, we propose a counterfactual method to estimate such missing observations and debias the existing supervisions. We validate the effectiveness of our counterfactual supervision on widely adopted image benchmark datasets: CUFED and PEC.