iReason: Multimodal Commonsense Reasoning using Videos and Natural Language with Interpretability

Abstract

Causality knowledge is vital to building robust AI systems. Deep learning models often perform poorly on tasks that require causal reasoning, which is often derived using some form of commonsense knowledge not immediately available in the input but implicitly inferred by humans. Prior work has unraveled spurious observational biases that models fall prey to in the absence of causality. While language representation models preserve contextual knowledge within learned embeddings, they do not factor in causal relationships during training. By blending causal relationships with the input features to an existing model that performs visual cognition tasks (such as scene understanding, video captioning, video question-answering, etc.), better performance can be achieved owing to the insight causal relationships bring about. Recently, several models have been proposed that have tackled the task of mining causal data from either the visual or textual modality. However, there does not exist widespread research that mines causal relationships by juxtaposing the visual and language modalities. While images offer a rich and easy-to-process resource for us to mine causality knowledge from, videos are denser and consist of naturally time-ordered events. Also, textual information offers details that could be implicit in videos. We propose iReason, a framework that infers visual-semantic commonsense knowledge using both videos and natural language captions. Furthermore, iReason's architecture integrates a causal rationalization module to aid the process of interpretability, error analysis and bias detection. We demonstrate the effectiveness of iReason using a two-pronged comparative analysis with language representation learning models (BERT, GPT-2) as well as current state-of-the-art multimodal causality models.

arXiv comment12 pages, 1 figure, 7 tables

References in corpus (17)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• LXMERT: Learning Cross-Modality Encoder Representations from Transformers
• Causality for Machine Learning
• Women also Snowboard: Overcoming Bias in Captioning Models
• CausalGAN: Learning Causal Implicit Generative Models with Adversarial Training
• Recent Advances in Natural Language Inference: A Survey of Benchmarks, Resources, and Approaches
• Visual Causal Feature Learning
• From Images to Sentences through Scene Description Graphs using Commonsense Reasoning and Knowledge
• Causal Induction from Visual Observations for Goal Directed Tasks
• Unbiased Scene Graph Generation from Biased Training
• Spatio-Temporal Graph for Video Captioning with Knowledge Distillation
• iPerceive: Applying Common-Sense Reasoning to Multi-Modal Dense Video Captioning and Video Question Answering
• Causal Discovery from Incomplete Data: A Deep Learning Approach
• More Grounded Image Captioning by Distilling Image-Text Matching Model
• Two Causal Principles for Improving Visual Dialog
• Joint Reasoning for Temporal and Causal Relations
• Learning Contextual Causality from Time-consecutive Images