Quantum computers currently have limited computational capacity, with the most powerful quantum computer containing only a few hundred qubits, while the chip in a smartphone has billions of transistors. Additionally, they suffer from unreliability, producing different results for the same calculation when run repeatedly.
However, due to their inherent ability to simultaneously consider multiple possibilities, quantum computers can address certain complex computational problems even with their relatively small size. In a recent development, IBM researchers have announced a method to effectively manage this unreliability, resulting in reliable and useful answers.
Dorit Aharonov, a computer science professor at the Hebrew University of Jerusalem, commended IBM's achievement as a significant step towards advancing quantum algorithmic design. While Google claimed "quantum supremacy" in 2019 by performing a task much faster on a quantum computer than on a classical one, IBM's researchers have achieved something novel and more practically valuable, although it is referred to with a more modest name.
Jay Gambetta, a vice president at IBM Quantum, described this phase of quantum computing as the "era of utility" when quantum computers can provide practical value. The findings of the IBM team, led by Dr. Gambetta, were published in the journal Nature.
Traditional computers, known as classical or digital computers, operate with bits that represent either a 1 or a 0. In contrast, quantum computers use qubits, which can represent a more complex state of information. Similar to the famous Schrödinger's cat thought experiment, where a cat could exist in both a dead and alive state simultaneously, a qubit can be both 1 and 0 at the same time. This property enables quantum computers to perform multiple calculations simultaneously, whereas classical computers must perform each calculation sequentially. Quantum computers have the potential to solve complex problems in fields like chemistry and materials science that are currently beyond the reach of classical computers. However, they also raise concerns about privacy as quantum algorithms could compromise existing encryption and password protection methods.
Although Google's claim of quantum supremacy faced skepticism from IBM and others, the IBM researchers conducted a different task of interest to physicists. They used a quantum processor with 127 qubits to simulate the behavior of 127 atom-scale bar magnets, governed by the rules of quantum mechanics, in a magnetic field. This simulation focused on the Ising model, a basic system for studying magnetism.
The calculation on the quantum computer was completed in less than a thousandth of a second. Each individual quantum calculation was unreliable due to quantum noise, but their speed allowed for repeated computations. In fact, intentional additional noise was introduced to make the answers even more unreliable. By analyzing the specific characteristics and effects of the noise at each step of the calculation, the researchers were able to mitigate the errors caused by noise and obtain accurate results. They referred to this process as error mitigation.
The researchers performed the calculation 600,000 times, converging on an answer for the overall magnetization produced by the 127 bar magnets. To validate their results, the IBM team collaborated with physicists at the University of California, Berkeley. Although the Ising model with 127 bar magnets is too large for conventional computers to handle precisely, classical algorithms can provide approximate solutions, similar to how JPEG compression discards less crucial data to reduce file size while preserving most details of an image.
Initially, the Berkeley physicists expected their classical algorithms to outperform the quantum ones. However, in cases where the quantum and classical calculations diverged and exact solutions were unknown, the quantum results proved to be more accurate. While it remains uncertain whether quantum computing is superior to classical techniques for the Ising model, researchers are exploring error correction as a long-term solution to detect and rectify calculation errors, potentially paving the way for quantum computers to excel in various applications.
IBM's scientists believe IBM's scientists believe that error mitigation serves as an interim solution for tackling increasingly complex problems beyond the Ising model. They consider it a starting point to explore more intriguing challenges in the realm of natural sciences.
These endeavors could involve unraveling the properties of exotic materials, expediting drug discovery processes, and modeling fusion reactions. By pushing the boundaries of quantum computing and expanding its applicability to diverse scientific domains, researchers aim to harness the immense potential of quantum computers to revolutionize various fields of study and practical applications.