Leeseok Kim's graduate work couldn't have come at a better time. The Ph.D. student at the University of New Mexico Department of Electrical and Computer Engineering has discovered something that could finally crack one of quantum computing's most stubborn problems: noise. In research published in Physical Review Letters and presented at QSim 2025, an international conference on quantum simulation, Kim and his team have shown that injecting randomness into quantum control protocols can outperform the deterministic methods currently protecting quantum computers from their own failures.
This matters because quantum computers promise to revolutionize how we solve certain classes of problems — from simulating and discovering new materials to tackling complex optimization challenges and encryption. Yet despite decades of development, these machines remain frustratingly fragile. Quantum systems are exquisitely sensitive; any stray electromagnetic field, vibration, or thermal fluctuation introduces noise that corrupts calculations and renders results unreliable. Building quantum computers powerful enough to solve practical problems at scale has remained one of science's great unsolved challenges, largely because controlling noise is so difficult.
Kim's breakthrough, developed under the guidance of Assistant Professor Milad Marvian with support from Changhao Yi, takes a counterintuitive approach. Rather than trying to eliminate randomness, the team harnessed it. They developed a new randomized construction of dynamical decoupling — a quantum control technique already widely used in modern devices — and proved mathematically that their randomized version can outperform any deterministic counterpart, including the noise-suppression protocols currently deployed in quantum machines around the world.
"In this project, we found a way to use randomized strategies to control quantum systems to outperform existing robust control protocols," Kim explained. What makes this discovery particularly elegant is its practicality. The approach is simple enough that it can be integrated into existing quantum computing platforms without major overhauls. Researchers won't need to redesign their hardware or rebuild their systems — they can apply this technique to devices already in the field, making adoption straightforward.
The implications ripple outward. If quantum computers can more effectively suppress noise using Kim's method, they can run longer, more reliable calculations. That moves the needle toward quantum advantage in real-world applications, from drug discovery to materials science to artificial intelligence. Marvian, reflecting on Kim's work, noted that his "Ph.D. thesis, which received distinction, advances our understanding of a range of important problems, from noise suppression to the power of quantum computers in simulation tasks."
Kim earned his bachelor's degree from Cornell University and will transition directly into a postdoctoral fellowship at Los Alamos National Laboratory this spring — a move that underscores how seriously the quantum computing community is taking this work. The fact that a graduating Ph.D. student from a university research group has produced findings significant enough to shape next-generation quantum computing platforms speaks to the quality of research happening at institutions beyond the usual tech giants. As quantum computing moves from theoretical promise toward practical reality, innovations like Kim's — modest in presentation, profound in potential — are what will eventually bridge the gap between what these machines can theoretically do and what they can actually deliver.