An Exploratory Study of AI System Risk Assessment from the Lens of Data Distribution and Uncertainty
This addresses the problem of assessing AI system risks for practitioners, but it is incremental as it builds on existing model-level studies.
The paper tackles the lack of systematic risk assessment for AI systems by conducting an exploratory study from data distribution and uncertainty perspectives, resulting in key findings from large-scale experiments (700+ configurations, 5000+ GPU hours) that highlight practical needs for deeper investigation.
Deep learning (DL) has become a driving force and has been widely adopted in many domains and applications with competitive performance. In practice, to solve the nontrivial and complicated tasks in real-world applications, DL is often not used standalone, but instead contributes as a piece of gadget of a larger complex AI system. Although there comes a fast increasing trend to study the quality issues of deep neural networks (DNNs) at the model level, few studies have been performed to investigate the quality of DNNs at both the unit level and the potential impacts on the system level. More importantly, it also lacks systematic investigation on how to perform the risk assessment for AI systems from unit level to system level. To bridge this gap, this paper initiates an early exploratory study of AI system risk assessment from both the data distribution and uncertainty angles to address these issues. We propose a general framework with an exploratory study for analyzing AI systems. After large-scale (700+ experimental configurations and 5000+ GPU hours) experiments and in-depth investigations, we reached a few key interesting findings that highlight the practical need and opportunities for more in-depth investigations into AI systems.