Quantum computing
What is quantum computing?
Quantum computers process information using qubits, which can be in superpositions of 0 and 1, in place of classical bits that represent either 0 or 1. By using entanglement between qubits and interference between quantum states, they address certain problems through principles different from those of classical computers.
Quantum computing includes gate-based approaches that combine quantum gates for a broad range of computations, as well as approaches tailored to particular problems, such as quantum annealing and boson sampling. Current devices are limited in qubit count, computational accuracy, and circuit depth. Researchers are investigating what can be computed within these constraints.
How can we use quantum computers?
It is not always clear which problems benefit from quantum computing. Researchers are therefore studying how quantum circuits of limited size can be applied to tasks such as machine learning and optimization.
Examples include quantum machine learning, in which circuit parameters are trained to classify data, and combinatorial optimization, which searches for good combinations among many candidates. At Takeoka Lab, we study the Quantum Approximate Optimization Algorithm (QAOA), variational quantum algorithms, and quantum classifiers to understand the conditions under which quantum computation can be useful compared with classical methods.
Given the limitations of available quantum devices, we also study what can be computed with a small number of qubits or shallow quantum circuits. We investigate this through theory, numerical simulations, and collaborative experiments with photonic quantum circuits.
Scaling up quantum computing
Improving quantum computers requires more qubits as well as precise control of quantum states while keeping errors low. Researchers study noise in real devices to inform quantum error correction, and develop quantum optimal control methods for manipulating quantum systems precisely.
Distributed quantum computation offers an alternative to enlarging a single quantum computer: multiple computers or modules are connected through a network and work together. If entanglement can be generated and quantum information exchanged between different devices, they may be used as a single computing resource.
We study optical connections between quantum computers, quantum optimal control, quantum error correction, and continuous-variable photonic quantum information processing. Drawing on both quantum optics and quantum information theory, we consider physical devices and information processing together.
Quantum computing research at Takeoka Lab
Quantum computing is one part of our research. We have expanded into this area by drawing on ideas developed through our work in quantum optics and information theory. We currently study how to scale quantum computing by connecting computers, and how to apply it to tasks such as machine learning and optimization.
See the Distributed Quantum Computation and Quantum Machine Learning team pages for research topics and members.
Distributed Quantum Computation
We study technologies that connect quantum computers, using light and other means, so that they can compute together. Our work covers a range of challenges in scaling up quantum computing, including connections between quantum nodes, quantum optimal control, quantum error correction, and continuous-variable photonic quantum information processing.
Distributed Quantum Computation
Quantum Machine Learning
We study how quantum circuits can be used for machine learning and optimization. Through quantum classifiers, variational quantum algorithms, and combinatorial optimization, we use theory and numerical simulations to investigate which computations can make use of current quantum devices.