×

IBM's experiment hints at quantum computers surpassing classical ones in 2 years

Eagle quantum computer demonstrates faster simulation of real material properties

In a groundbreaking experiment conducted by IBM, quantum computers have shown promising signs of outperforming classical digital computers in practical tasks. This development has raised the possibility that true quantum supremacy, where quantum computers surpass classical ones, could be realized sooner than anticipated.

Quantum computers derive their power from exploiting the principles of quantum mechanics. At the quantum level, matter exhibits both particle and wave properties, and quantum computing capitalises on this behaviour through specialised hardware. Unlike classical physics, which fails to explain the operation of these quantum devices, a scalable quantum computer holds the potential to perform certain calculations exponentially faster than any modern classical computer.

Classical computers, including smartphones and laptops, store information in binary bits that represent either 0 or 1. Conversely, quantum computers utilise quantum bits, or qubits, as the fundamental units of memory. Here lies the edge of quantum computers over classical ones. In scenarios with a vast number of potential combinations, quantum computers can consider them simultaneously. Examples include solving complex problems such as finding prime factors of large numbers or determining the optimal route between two locations.

However, it is important to note that classical computers may still excel in certain situations, implying that future computers might harness a combination of both quantum and classical computing technologies.

Currently, quantum computers are highly sensitive to external factors, such as heat, electromagnetic fields, and interactions with air molecules, which can cause a qubit to lose its quantum properties. This phenomenon, known as quantum decoherence, leads to system failures, which occur more rapidly as the number of particles involved increases.

To safeguard qubits from external interference, quantum computers necessitate physical isolation, cooling mechanisms, and controlled energy pulses. Additionally, additional qubits are required to rectify errors that may arise within the system.

The recent study, published in the journal Nature, showcased the prowess of IBM's quantum computer, named Eagle, in simulating the magnetic properties of a real material at a faster rate than a classical computer. The achievement was made possible by implementing a specialized error-mitigation process that compensated for noise, a fundamental vulnerability of quantum computers.

Historically, quantum computers have faced a significant challenge: the delicate quantum states of qubits are susceptible to even the slightest disruption from the external environment, compromising the information they carry. Consequently, quantum computers have been prone to errors and characterized as "noisy."

It is worth noting that any computational problem solvable by a classical computer can also be solved by a quantum computer, and vice versa, adhering to the Church-Turing thesis. However, quantum algorithms for specific problems exhibit significantly lower time complexities than their classical counterparts. Notably, quantum computers have the potential to solve problems that are practically infeasible for classical computers in a reasonable timeframe, a concept known as "quantum supremacy."

The successful proof-of-principle experiment involved IBM's 127-qubit Eagle supercomputer, built on superconducting circuits. By measuring the noise generated by each qubit and predicting its impact, the researchers effectively simulated the magnetic state of a two-dimensional solid material.

While previous claims of quantum supremacy have emerged, such as Google's 2019 experiment with its Sycamore quantum computer, which lacked practical applications, the IBM demonstration tackled a real, albeit simplified, physical problem. This breakthrough inspires optimism that the approach could be extended to other systems and more complex algorithms, as noted by John Martinis, a physicist at the University of California, Santa Barbara.