IBM Quantum is excited to announce the second annual Open Science Prize — an award for those who can present an open source solution to some of the most pressing problems in the field of quantum computing. Submissions are open now, and must be received by April 16, 2022.
This year, the challenge will feature one problem from the field of quantum simulation, solvable through one of two approaches. The best open source solution to each approach will receive a $40,000 prize, and the winner overall will receive another $20,000.
Simulating physical systems on quantum computers is a promising application of near-term quantum processors. This year’s problem asks participants to simulate a Heisenberg model Hamiltonian for a three-particle system on IBM Quantum’s 7-qubit Jakarta system. The goal is to simulate the evolution of a known quantum state with the best fidelity as possible using Trotterization.
Researchers use the Heisenberg model to study a variety of physical systems involving interacting particles with spins. Quantum computers are useful tools to simulate these models because you can represent the spin states of particles as the computational states of qubits.
But tackling this Hamiltonian can prove challenging, since different subsets of qubits in the same system don’t commute — that is, you can’t measure subsets of the problem simultaneously to a high precision, due to the The Uncertainty Principle: In quantum physics, we cannot simultaneously know two non-commuting variables (like the position and momentum of a particle). This implies that a quantum system in a perfectly definite state can be certain under one measurement and completely random under another. Moreover, if a quantum system starts out in an arbitrary unknown state, no measurement can reveal complete information about that state; the more information the measurement reveals, the more the state is disturbed. This is a underlying principle of quantum cryptography. Source: Qiskit GlossaryUncertainty Principle.
Trotterization allows us to simulate these kinds of systems by quickly switching between the non-commuting parts.
We picked this problem because the Heisenberg model is ubiquitous and relatively simple — therefore, it’s a great place to start for those just dipping a toe into quantum simulation. But also, given the model’s ubiquity, any solution that betters our ability to simulate it will have broad impact on the field of quantum simulation overall.
Team up to win
Participants can team up into groups of up to five, and can choose to solve the problem in one of two ways:
- Either use Qiskit Pulse, that is, the Qiskit module that allows users pulse-level control over quantum quantum gates,
- Or try to solve the problem using Qiskit defaults.
We encourage each team to push outside of their members’ comfort zone and try whichever method they think is best suited to solve the problem. Although Qiskit Pulse offers more detailed control of the qubits, there are advantages and disadvantages to both approaches.
Closing the door on last year’s prize
Last year’s The IBM Quantum team selected four winning teams for last year’s Graph State Challenge, and acknowledged a creative solution to the SWAP gate challenge. Read about the results.Open Science Prize was a hit, as participants used Qiskit Pulse to simulate two challenging outstanding problems whose solutions could help advance the field of quantum computation. We had more than 30 submissions to two cutting edge research challenges that were, and still are, open questions in the field. Participants used Qiskit to run more than 6 billion circuits on IBM Quantum’s 7-qubit Casablanca system. We hope to see even more participation this year from across the quantum computing community.