FENG Ling

Dr. FENG Ling is an Adjunct Assistant Professor at the National University of Singapore and Manager of the Complex System Group at the Institute of High Performance Computing, A*STAR. His work bridges theoretical and applied research in complex systems, focusing on statistical physics principles underlying deep learning and phase transitions in neural networks, as well as percolation phenomena in inter-dependent networks.

• Education: PhD in Physics, National University of Singapore (2013)

His research spans complexity science, artificial neural networks, and social/economic systems. By analyzing critical states between periodic cycles and chaos, he explores optimal neural network training and explainability. He also develops frameworks for systemic spreading in complex networks, applicable to disease propagation, information diffusion, and blockchain dynamics.

The selected publications highlight his contributions to understanding 1/f noise in deep neural networks (2024), reconstructing networked complex systems (2024), and edge-of-chaos training principles (2024). Earlier works investigate generalization in deep learning (2020), viral spreading on social networks (2018), and global spreader identification (2018), reflecting interdisciplinary applications of percolation theory and machine learning.

Scientific Awards

• 9th place in Predicting Generalization in Deep Learning Competition at NeurIPS 2020

Dr. Feng leads the Complex System Group at A*STAR's Institute of High Performance Computing, where his team develops algorithms for maximizing or mitigating systemic spread in social, financial, and blockchain networks. His work integrates nonlinear dynamics, computational modeling, and data science to address challenges in artificial intelligence and networked systems.