Attentive Explanations: Justifying Decisions and Pointing to the Evidence

Abstract

Deep models are the defacto standard in visual decision models due to their impressive performance on a wide array of visual tasks. However, they are frequently seen as opaque and are unable to explain their decisions. In contrast, humans can justify their decisions with natural language and point to the evidence in the visual world which led to their decisions. We postulate that deep models can do this as well and propose our Pointing and Justification (PJ-X) model which can justify its decision with a sentence and point to the evidence by introspecting its decision and explanation process using an attention mechanism. Unfortunately there is no dataset available with reference explanations for visual decision making. We thus collect two datasets in two domains where it is interesting and challenging to explain decisions. First, we extend the visual question answering task to not only provide an answer but also a natural language explanation for the answer. Second, we focus on explaining human activities which is traditionally more challenging than object classification. We extensively evaluate our PJ-X model, both on the justification and pointing tasks, by comparing it to prior models and ablations using both automatic and human evaluations.

References in corpus (3)

• Microsoft COCO Captions: Data Collection and Evaluation Server
• Dynamic Memory Networks for Visual and Textual Question Answering
• Towards Transparent AI Systems: Interpreting Visual Question Answering Models

Cited by in corpus (19)

• A Survey on the Explainability of Supervised Machine Learning
• Visual Entailment: A Novel Task for Fine-Grained Image Understanding
• Levels of explainable artificial intelligence for human-aligned conversational explanations
• Towards Interpretable Deep Neural Networks by Leveraging Adversarial Examples
• Interpretable to Whom? A Role-based Model for Analyzing Interpretable Machine Learning Systems
• Understanding Attention and Generalization in Graph Neural Networks
• Using Machine Learning Safely in Automotive Software: An Assessment and Adaption of Software Process Requirements in ISO 26262
• It Takes Two to Tango: Towards Theory of AI's Mind
• Embedding Deep Networks into Visual Explanations
• Trepan Reloaded: A Knowledge-driven Approach to Explaining Artificial Neural Networks
• Recent Advances and Trends in Multimodal Deep Learning: A Review
• The What, the Why, and the How of Artificial Explanations in Automated Decision-Making
• Interpretable Discovery in Large Image Data Sets
• Attentive Explanations: Justifying Decisions and Pointing to the Evidence (Extended Abstract)
• Understanding Convolutional Networks with APPLE : Automatic Patch Pattern Labeling for Explanation
• Grounding Visual Explanations (Extended Abstract)
• Interpretable Visual Reasoning via Induced Symbolic Space
• Explainable Neural Computation via Stack Neural Module Networks
• Assisting human experts in the interpretation of their visual process: A case study on assessing copper surface adhesive potency