Yingdi Wang

Papers

Total Citations

H-Index

About

Yingdi Wang is a researcher focused on the intersection of artificial intelligence and cybersecurity, with a particular emphasis on adversarial machine learning and its implications for autonomous systems. Their most cited work, "A Training-Based Identification Approach to VIN Adversarial Examples in Path Planning" (2021), addresses a critical vulnerability in AI-driven path planning: the susceptibility of neural networks to adversarial examples—subtly altered inputs that cause erroneous outputs. Wang proposes a training-based identification method to detect such attacks, contributing to the robustness of visual inertial navigation (VIN) systems used in robotics and autonomous vehicles. Though early in their career, with this paper accruing 2 citations, Wang’s research tackles a pressing challenge in AI security, where even minor perturbations can lead to catastrophic failures in real-world applications. Their work underscores the importance of developing resilient algorithms for safety-critical domains. As adversarial threats evolve, Wang’s identification approach offers a foundational step toward more trustworthy AI, positioning them as a rising voice in the field of trustworthy machine learning and cyber-physical system security.