IBM's QC innovation roadmap will demo tech needed to realize error correction based on novel Gross code: Venkat Subramaniam

There has been lot of talk around quantum computing. Various companies are on their way towards developing sustainable QC systems. The idea is to help solve challenges that classical computing has not been able to do, so far.

Venkat Subramaniam, Quantum Lead, IBM Research India, tells us more about what IBM has been doing in this field. Excerpts from an interview:

DQ: Do you think utilizing quantum computing in the cloud is a practical choice for the majority of businesses?

Venkat Subramaniam: Cloud access to quantum systems gives organizations a practical way to explore how the technology could help solve their industry’s challenges that classical computers have not been able to. At IBM, our quantum ecosystem includes more than 500,000 registered users and more than 250 organizations — from individual developers to domain experts at universities, Fortune 500s, government labs, and startups. And they all have access, over the cloud, to our latest utility-scale quantum systems, via different access plans.

DQ: By when can an industry around QC software and algorithms truly arrive?

Venkat Subramaniam: Today, IBM provides the full stack of quantum hardware, services, and software — including the recently announced Qiskit 1.0, the world’s most widely used open-source quantum programming software. It will provide programming with Qiskit Patterns — a four-step process for running algorithms on a quantum computer — plus, a new set of tools using AI to help write and optimize code.

Now, we are able to demonstrate how generative AI, available through watsonx (IBM's enterprise AI platform), can help automate the development of quantum code for Qiskit by fine-tuning the IBM Granite 20-billion parameter code foundation model.

Additionally, the IBM Quantum Network includes more than 250 organizations, including quantum software companies and companies and institutions developing quantum algorithms -- including IIT Madras and BosonQ Psi from India.

For example, the University of Tokyo, which recently installed a utility-scale 127-Qubit IBM Quantum Eagle processor in their IBM Quantum System One, will use the system to augment its pursuit of research interests in bioinformatics, materials science, and finance.

DQ: By when can we use QCs for simulating complex systems and behaviors in near real-time, and with high fidelity?

Venkat Subramaniam: In June of this year, our scientists demonstrated with UC Berkeley that quantum systems with more than 100 qubits are now capable of serving as scientific tools to explore utility-scale classes of problems. In the experiment, our team used the 127-qubit IBM Quantum 'Eagle' quantum processor to generate large, entangled states that simulate the dynamics of spins in an Ising model, and accurately predict properties such as its magnetization.

This utility-scale experiment showed that quantum computers are now tools capable of performing reliable computations at a scale beyond brute force classical computing methods that provide exact solutions to computational problems.

At this month’s IBM Quantum Summit, we announced our new processor, IBM Heron. IBM Heron’s 133 fixed-frequency qubits’ new architecture with tunable couplers yields a three- to five-fold improvement in error rates over our best 127-qubit IBM Eagle processors — the processor used in the successful quantum-utility experiment with UC Berkeley. In addition, Heron’s tunable coupler architecture virtually eliminates cross-talk associated with running simultaneous two-qubit gates.

As mentioned previously, all IBM quantum systems over the cloud are powered by utility-scale processors with more than 100 qubits.

DQ: What are the future trends for quantum computing in 2024 and beyond?

Venkat Subramaniam: At the IBM Quantum Summit, we expanded our industry-defining roadmap out to 2033 for a decade worth of quantum innovation. The future processors, shown on our 10-year roadmap, are intended to improve the quality of operations they will be capable of running — from today’s utility-scale systems, to systems that significantly extend the complexity and size of workloads they are capable of handling.

For our users, this means allowing more advanced utility-scale work and a frictionless development environment on this long-term journey toward quantum-centric supercomputing.

In terms of the specific technology detailed in the roadmap, we highlight improvements in the number of gates that our processors and systems will be able to execute. Starting next year, 2024, with a target of Heron reaching 5,000 gates, the roadmap lays out multiple generations of processors, each leveraging improvements in quality to achieve ever-larger gate counts.

Our new innovation roadmap will demonstrate the technology needed to realize error correction based on novel Gross (a dozen dozen) code — developed by our scientists — through l-, m-, and c-couplers to be demonstrated by Flamingo, Crossbill processors in 2024, and the Kookaburra processor in 2025.

In 2029, we will hit an inflection point: executing 100 million gates over 200 qubits with our Starling processor employing the error-correction Gross (a dozen dozen) code. The Starling processor will be followed by a Blue Jay processor-powered system capable of executing 1 billion gates across 2,000 qubits by 2033. This represents a nine order-of-magnitude increase in performed gates since we put our first device on the cloud in 2016.

While error correction is far from a solved problem, the new Gross code and other advances across the field are increasing our confidence that fault-tolerant quantum computing isn’t just possible, but is possible without having to build an unreasonably large quantum computer.