Opening New Frontiers in Information Science with Quantum Theory
When quantum theory was established around 1925, it was often viewed as “a framework that explains phenomena but is too counterintuitive to understand what is really happening.” Today, quantum theory is widely understood as an unusual and counterintuitive probabilistic theory. From this viewpoint, quantum information science has emerged as the study of information processing governed by quantum rules.
In the quantum world, the behavior of information itself changes: for example, information can be stored in superposition until measurement, yet it cannot be copied. As a result, tasks that once seemed impossible can become feasible—such as ultrafast communication, ultraprecise sensing, and even potential algorithmic speedups for hard computational problems.
Our group tries to understand quantum theory as a theory of information, and study—at a theoretical level—the potential and fundamental limits of quantum information processing. Our work spans quantum communication, quantum error correction, quantum algorithms, and the behavior of information in quantum many-body systems.
Below we highlight some research topics we have worked on. Beyond these, whenever new questions or unexpected connections arise from the interplay between quantum theory and information, we are eager to expand our scope accordingly.
Quantum Pseudorandomness
Randomness is a key resource in information science. However, ideal randomness is rarely available in practical settings as it is inherently hard to generate. This makes it crucial—both theoretically and practically—to ask how well randomness can be approximated under limited computational resources.
Card shuffling is a great example: how many shuffles are needed before a deck is “well mixed”? One can show that about seven shuffles already make the deck statistically close to random. In other words, one can obtain “essential randomness” with surprisingly few operations—without achieving perfect randomness.
Quantum pseudorandomness is the quantum analogue of this question. Our goal is to understand how far ideal quantum randomness can be approximated by physically and computationally feasible quantum dynamics. By studying efficient mechanisms to generate quantum pseudorandomness and developing theoretical tools to certify its properties, we aim to clarify what “practical quantum randomness” really means.
Keywords: Haar randomness; unitary/state designs; random quantum circuits; random quantum dynamics; computational complexity of quantum randomness; Weingarten calculus
Quantum Communication and Quantum Error Correction
How reliably can one transmit information? This central question in information science is far more challenging in the quantum setting as quantum states are highly sensitive to noise and loss. Protecting quantum information is therefore essential, and the basic framework for doing so is quantum error correction.
Quantum error correction suppresses noise by encoding quantum information redundantly. Many quantum codes exist, and in particular, quantum random codes leverage quantum randomness and quantum pseudorandomness to achieve extremely high error-correction performance.
Introducing randomness makes a lot of sense in error correction. Consider boarding an airplane: if passengers board strictly by seat number, bottlenecks form at the overhead bins and the process slows down. If boarding is randomized to some extent, movement becomes distributed and seating proceeds more smoothly. Randomization avoids specific collisions on average and helps prevent worst-case congestion. In a similar spirit, incorporating randomness into quantum codes can reduce bias toward particular noise patterns and lead to robust error-correction performance.
In our group, we study a broad range of quantum codes, including random codes. We address fundamental questions such as “Why do quantum random codes work?” and “How much quantum randomness is needed?” At the same time, we also pursue stabilizer-code research aimed at balancing rigorous performance guarantees with practical implementability.
Keywords: quantum error-correcting codes; quantum random codes; low-depth general decoding circuits; quantum Shannon theory; entanglement distribution; decoupling theory
Quantum Algorithms and Quantum Complexity
Algorithms form the foundation of all information processing. In 1994, Shor’s discovery of a polynomial-time quantum algorithm for integer factorization triggered rapid growth in quantum information science.
Early quantum algorithm research often relied on problem-specific, highly specialized techniques. More recently, the advent of a powerful subroutine known as quantum singular value transformation has provided a unifying framework for algorithmic approaches across a wide range of problems.
Building on this development, our group revisits the topics above from an algorithmic perspective. We have worked on quantum algorithms to generate and certify quantum pseudorandomness, on decoding algorithms for quantum random codes, and on implementing core primitives from quantum information theory—such as the Uhlmann transformation—as explicit quantum algorithms. Constructing algorithms not only makes these protocols executable in principle, but also enables a concrete analysis of the computational resources they require.
As scalable quantum computing becomes increasingly realistic, the algorithmic viewpoint is more important than ever. Our goal is to leverage algorithmic approaches to make all quantum protocols realicstically implementable.
Keywords: quantum algorithms; quantum complexity; quantum query complexity; quantum singular value transformation; block-encoding; algorithmic quantum information theory
New Physics Through the Lens of Information
Quantum information science sits at the intersection of information science and physics. Information science underpins our modern, information-driven society, while quantum theory is a cornerstone of modern physics. Bringing these fields together opens new perspectives—and new research questions.
For instance, consider three notions: “error correction” as a technology to protect information, “chaos” as unpredictable dynamics, and “thermalization” as the approach to thermal equilibrium. They may seem unrelated, yet they are deeply connected.
A useful intuition comes from random codes. By randomization, random coding suppresses bias and can achieve strong error correction. Doing so typically requires encoding by “sufficiently random and complex” dynamics. The key point is that, precisely because there is no bias, the resulting behavior looks completely unpredictable to anyone who does not know the code. This naturally bridges error correction and chaos. Moreover, if one observes only a subsystem after such dynamics, the behavior becomes thermalized. In this sense, “random codes,” “chaos,” and “thermalization” are different faces of the same underlying principle—unpredictable dynamics—seen through the lens of information and physics.
This cross-disciplinary viewpoint is powerful because it generates new ideas almost automatically: for example, using chaotic dynamics as a resource for information processing, or understanding everyday thermalization as a form of information encoding. A small shift in perspective can change how the world looks. Our group also explores this line of thinking at the boundary between information and physics.
Keywords: quantum many-body systems; quantum chaos; thermalization; information scrambling; statistical mechanics; complex systems