Summary: IBM is on its way to error-tolerant quantum computing, but before we get to this holy grail of quantum computing, there is a lot of useful work that can be done using error-mitigating techniques.
I recently had the opportunity to visit IBM’s Quantum Research Laboratories in Yorktown Heights, New York to speak with Jay Gambetta, IBM Fellow and Vice President of Quantum Computing, IBM Research and his team working on quantum computing development. IBM is making steady progress on its quantum roadmap (ibid.) but this is still a nascent technology and there is still a lot of experimentation needed to advance the quantum computing market, which is why the research is so important.
The goal for IBM researchers is to make quantum computing as ubiquitous as possible to solve unique problems. To make quantum systems more accessible, they must become “cloud-native” or “serverless”, becoming a usage-charged cloud resource. In this age of classified data centers, quantum can be one of the specialized computing elements available to classic computers, just like today’s GPUs.
With the goal of scaling quantum systems to more than a million qubits, IBM Research is following a path similar to that taken with classical computers: putting more and faster qubits on a chip using a silicon scale; Connect many quantum dies like tiles; and build groups of quantum computers working together. I’ve previously written about IBM’s roadmap to building systems with more qubits and linking multiple quantum systems together.
The cryostat (the chamber that cools the quantum chip to near absolute zero) also needs to get larger to accommodate larger chips with greater I/O. IBM has teamed up with Blue Force to help build an essential ecosystem for System 2 large refrigerants.
IBM is also increasing the density of its cooled infrastructure for RF signal input and output by leveraging commercial technology.
Quantum computing’s journey to quantum advantage
While the goal is to build systems with millions of raw qubits for error-tolerant quantum computing, there is a lot of work that can be done in the meantime to improve the performance of raw qubits to do more work sooner using quantum error mitigation, as shown in Figure below.
To obtain better quantum results using relatively noisy and short-lived qubits, some alternative solutions are required. IBM Research has come up with two proven error mitigation techniques. Current quantum devices are subject to various sources of noise degradation. This includes qubit decoherence, single gate errors, and scaling errors. These issues limit the number of phases that can be implemented in a quantum circuit today. Even shallow circuits can be subjected to noise which can lead to wrong estimates. For a deeper discussion about error reduction, IBM has published a file Blog post newly.
Noise mitigation techniques are very technical. From 2017 IBM Review Letter Error mitigation for short depth quantum circuits To the American Physical Society: “The first method, extrapolating to the zero noise limit, subsequently cancels out the noise disturbance forces by applying the maximally delayed Richardson approach. The second method eliminates errors by re-sampling the random circuits according to a quasi-probability distribution.” Like I said, it’s technical, but IBM researchers can hide the details inside the Qiskit Runtime environment.
The ultimate goal of practical quantum computing is to provide an advantage over classical computing for solving important problems in a reasonable time frame. The most obvious advantage is that the problem is solved in significantly less time. To achieve this, the problem must be represented as a quantum circuit and not simulated on a classical system, which means that quantum computers will not replace classical computers or even GPU computing but are here to solve a unique class of problems.
For a quantum computer to have an advantage over a classical computer (the so-called quantum advantage) requires mapping the problem to quantum circuits with better solutions than traditional methods and the ability to obtain reliable results with faster run times. IBM researchers work with industry partners to identify those problems that need better solutions.
To measure progress, IBM has a measure of the quality of qubits called quantum size (QV) and circuit velocity called circuit layer operations per second (CLOPS). This provides a more complete picture of the progress of quantum computing than just pure qubit numbers.
There is still much that can be achieved by blending classical and quantum computing. Through a process called circuit knitting, quantum circuits are broken down into smaller circuits and made use of classical computing to evaluate temporary outcomes. This could enhance the quantum workflow on the available qubits.
One application of the blending of classical and quantum computing is computational chemistry. To calculate the electron valence, the electron cloud can be divided into active and inactive parts. The passive cloud is calculated with classical computers and the active part of the cloud uses a quantum modeling method called density functional theory (DFT).
IBM is constantly working with partner companies to explore areas where quantum computing can make a difference in solving challenging problems.
Quantum Needs Programs
IBM Research is also building a software suite for quantum middleware, using lessons learned from classical computing and GPU arithmetic. IBM moved from a static language and added dynamic circuits, where the output of measurements taken in the middle of a circuit is used to identify future gates in the same circuit. The latest developments include support for conditional circuits in the open quantum development platform QASM3.
Programming challenges for quantum circuits include optimizing quantum circuit depth, finding alternative models, and adding parity checks for quantum error corrections. IBM also adds more function calls and basic fundamentals: the sampler and the estimator. These add-ons help shorten development times. Results Better accuracy and lower costs by reducing circuit run times.
The future of quantum computing is coming and a big part of the work is improving the quality of qubits, not just increasing the number of qubits. It takes a complete systems approach to building quantum systems with hardware, software middleware, and libraries. We also expect to see more interaction between quantum processing and AI processing in the near future.
Tirias Research tracks and consults companies across the electronics ecosystem from semiconductors to systems and sensors to the cloud. Tirias Research team members have consulted with IBM, Nvidia, Qualcomm, and other companies across AI and quantum systems.