Michael Wang

Michael Wang

Engineer. Researcher. Builder.

Publications

Journal Articles

BetaExplainer: A Probabilistic Method to Explain Graph Neural Networks

Published in Journal of Statistical Theory and Applications, 2025

BetaExplainer is a probabilistic GNN explanation framework that provides uncertainty estimates for edge importance and improves interpretability on complex graph datasets.

Recommended citation: Sloneker, W., Patel, S., Wang, H.-J., Crawford, L., & Singh, R. (2025). "BetaExplainer: A Probabilistic Method to Explain Graph Neural Networks." Journal of Statistical Theory and Applications, 24, 469–488.
Download Paper

Conference Papers

Time is of the Essence: Why Decision-Time Planning Costs Matter

Published in Finding the Frame: An RLC Workshop for Examining Conceptual Frameworks, 2024

Explored how decision-time planning costs influence reinforcement learning performance, highlighting when additional computation is worth the tradeoff.

Recommended citation: Wang, K. A., Xia, J., Chung, S., Wang, J., Piedrahita Velez, F., Wang, H.-J., & Greenwald, A. (2024). "Time is of the Essence: Why Decision-Time Planning Costs Matter." Finding the Frame: An RLC Workshop for Examining Conceptual Frameworks.
Download Paper