Timezone: »

 
Poster
Cost-Effective Interactive Attention Learning with Neural Attention Processes
Jay Heo · Junhyeon Park · Hyewon Jeong · Kwang Joon Kim · Juho Lee · Eunho Yang · Sung Ju Hwang

Thu Jul 16 08:00 AM -- 08:45 AM & Thu Jul 16 07:00 PM -- 07:45 PM (PDT) @
in Poster Session 43 »

We propose a novel interactive learning framework which we refer to as Interactive Attention Learning (IAL), in which the human supervisors interactively manipulate the allocated attentions, to correct the model's behaviour by updating the attention-generating network. However, such a model is prone to overfitting due to scarcity of human annotations, and requires costly retraining. Moreover, it is almost infeasible for the human annotators to examine attentions on tons of instances and features. We tackle these challenges by proposing a sample-efficient attention mechanism and a cost-effective reranking algorithm for instances and features. First, we propose Neural Attention Processes (NAP), which is an attention generator that can update its behaviour by incorporating new attention-level supervisions without any retraining. Secondly, we propose an algorithm which prioritizes the instances and the features by their negative impacts, such that the model can yield large improvements with minimal human feedback. We validate IAL on various time-series datasets from multiple domains (healthcare, real-estate, and computer vision) on which it significantly outperforms baselines with conventional attention mechanisms, or without cost-effective reranking, with substantially less retraining and human-model interaction cost.

Keywords: Healthcare · Safety · Time Series and Sequence Models · Accountability, Transparency and Interpretability

Author Information

Jay Heo (KAIST)
Junhyeon Park (KAIST)
Hyewon Jeong (KAIST)
Kwang Joon Kim (Yonsei University College of Medicine)
Juho Lee (AITRICS)
Eunho Yang (KAIST,AITRICS)
Sung Ju Hwang (KAIST, AITRICS)

More from the Same Authors