A Safety Assurable Human-Inspired Perception Architecture
This addresses safety-critical perception problems for autonomous applications, but it is incremental as it builds on existing human cognition models without presenting new empirical results.
The paper tackles the safety assurance limitations of deep neural network-based perception, such as adversarial vulnerability and non-interpretability, by proposing a dual process architecture inspired by human cognition, arguing it can systematically address these issues and enable human-level or better performance.
Although artificial intelligence-based perception (AIP) using deep neural networks (DNN) has achieved near human level performance, its well-known limitations are obstacles to the safety assurance needed in autonomous applications. These include vulnerability to adversarial inputs, inability to handle novel inputs and non-interpretability. While research in addressing these limitations is active, in this paper, we argue that a fundamentally different approach is needed to address them. Inspired by dual process models of human cognition, where Type 1 thinking is fast and non-conscious while Type 2 thinking is slow and based on conscious reasoning, we propose a dual process architecture for safe AIP. We review research on how humans address the simplest non-trivial perception problem, image classification, and sketch a corresponding AIP architecture for this task. We argue that this architecture can provide a systematic way of addressing the limitations of AIP using DNNs and an approach to assurance of human-level performance and beyond. We conclude by discussing what components of the architecture may already be addressed by existing work and what remains future work.