Hide-and-Seek: A Template for Explainable AI
arXiv:2005.00130
Abstract
Lack of transparency has been the Achilles heal of Neural Networks and their wider adoption in industry. Despite significant interest this shortcoming has not been adequately addressed. This study proposes a novel framework called Hide-and-Seek (HnS) for training Interpretable Neural Networks and establishes a theoretical foundation for exploring and comparing similar ideas. Extensive experimentation indicates that a high degree of interpretability can be imputed into Neural Networks, without sacrificing their predictive power.
24 pages, 14 figures. Submitted on a special issue for Explainable AI, on Elsevier's "Artificial Intelligence"
References in corpus (12)
- Distilling the Knowledge in a Neural Network
- Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
- Axiomatic Attribution for Deep Networks
- Learning Important Features Through Propagating Activation Differences
- Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation
- XLNet: Generalized Autoregressive Pretraining for Language Understanding
- Recurrent Models of Visual Attention
- Multiple Object Recognition with Visual Attention
- Sanity Checks for Saliency Maps
- Hierarchical Multiscale Recurrent Neural Networks
- A Theoretical Explanation for Perplexing Behaviors of Backpropagation-based Visualizations
- Improving Semantic Segmentation via Video Propagation and Label Relaxation