Dexiu Ma
Papers
1
Total Citations
13
H-Index
1
About
Dexiu Ma is a rising researcher in computational mathematics and neural dynamics, specializing in advanced recurrent neural network (RNN) models for solving time-dependent systems. Their most-cited work, a 2024 paper on pseudoinverse-free recurrent neural dynamics, tackles a critical gap in the field: prior RNN-based methods for time-dependent linear equations only constrained the variable itself, ignoring its derivatives. Ma’s innovation introduces a framework that simultaneously handles constraints on both the variable and its rate of change, enabling more robust and physically realistic solutions in real-time applications like robotics and control systems. With 13 citations already, this work signals strong early impact. Ma’s contributions are notable for pushing beyond conventional pseudoinverse-based approaches, reducing computational overhead while maintaining accuracy. Their research holds promise for advancing dynamic system solving, particularly where derivative constraints are essential—a frontier with growing relevance in engineering and applied mathematics. As a developing scholar, Ma is establishing a reputation for rigorous, problem-driven methodology that bridges theory and practical constraint handling.
Research Focus
Key Achievements
Top Papers
- 1