MAEA: Multimodal Attribution for Embodied AI

Vidhi Jain, Sahiti Yerramilli, Jayant Sravan Tamarapalli, Yonatan Bisk

Keywords: Interpretable Machine Learning, Embodied AI, Multimodal

TL;DR: We present MAEA, a framework to compute global attributions per modality of any differentiable EAI policy.

Abstract: Understanding multimodal perception for embodied AI is an open question because such inputs may contain highly complementary as well as redundant information for the task. A relevant direction for multimodal policies is understanding the global trends of each modality at the fusion layer. To this end, we disentangle the attributions for visual, language, and previous action inputs across different policies trained on the ALFRED dataset. Attribution analysis can be utilized to rank and group the failure scenarios, investigate modeling and dataset biases, and critically analyze multimodal EAI policies for robustness and user trust before deployment. We present MAEA, a framework to compute global attributions per modality of any differentiable policy. In addition, we show how attributions enable lower-level behavior analysis in EAI policies through two example case studies on language and visual attributions.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2307.13850/code)