Researchers Get Past the “Postselection Wall” to See the Elusive Phase Transition in a Quantum Development
Using universal quantum gates, scientists have reported the first postselection-free experimental observation of a measurement induced phase transition (MIPT), marking a major advancement in the area of quantum information science. According to a Communications Physics research, this major advance removes an obstacle in the study of quantum system evolution under continual observation.
Under the direction of Xiaozhou Feng, Jeremy Côté, Stefanos Kourtis, and Brian Skinner, the research team avoided the exponential complexity that is usually involved in these studies by using a unique tree-structured circuit architecture. They have produced a clear experimental map of a transition that was previously mostly limited to theoretical models by executing their protocol on a cutting-edge trapped-ion quantum computer.
You can also read Why Microwave Qubits Dominate the Quantum Computing
The Tug-of-War: Unitary Evolution vs. Measurement
MIPT, or Measurement Induced Phase Transition, is the key to the finding. There exist two opposing forces in a monitored many-body quantum system. Unitary operators, the fundamental components of quantum gates, operate on one side to jumble data and establish enormous networks of entanglement between qubits. Physicists refer to this as a “volume-law” phase, in which entanglement increases in direct proportion to the system’s size.
However, measurement also has the effect of collapsing these quantum states. The qubits are “reset” if measurements are made often enough or strongly enough to disrupt the entanglement and push the system into a “disentangling” or “area-law” phase.
Building durable quantum memories and understanding the basic boundaries of quantum computing depend on knowing where this tipping point is, the key value at which the system transitions from a highly entangled state to a disentangled one.
Shattering the “Postselection Wall”
The “postselection problem” has made it notoriously difficult to observe this shift experimentally to date. The exact same set of measurement results must often be reproduced by researchers to study the entanglement of a system. Because quantum measurements are probabilistic, as the system gets bigger, the likelihood of receiving the identical “measurement record” twice decreases rapidly. “Experimental realization of the Measurement Induced Phase Transition is particularly difficult,” the scientists say in their abstract, adding that the number of measurements needed usually increases uncontrolled.
To address this, the researchers built tree-shaped quantum circuits in place of conventional grid-like circuits. This design uses “weak” measurements after Haar-random unitary operators for each node. Because of this particular shape, a highly efficient classical decoding procedure is possible, scaling linearly as opposed to exponentially with the number of qubits.
You can also read Benchmarking MedMNIST dataset on real Quantum Hardware
Success on Trapped-Ion Hardware
Using the H1-1 trapped-ion quantum computer from Quantinuum, the researchers tested their idea. This study employed universal gates instead of “Clifford” gates, a limited set of processes that are simpler for classical computers to imitate, as was the case in many earlier tests.
The outcomes were spectacular. The volume-law phase (seen at low measurement rates) and the area-law phase (observed at high measurement rates) were clearly distinguished in the experimental data. The results, which are frequently used to “clean up” noisy data from existing quantum gear, were remarkably accurate in matching the team’s theoretical expectations, negating the need for error mitigation strategies.
You can also read UCoMP Breakthrough In Monitoring Dynamic Solar Activity
Why It Matters
The experiment was made simpler by the tree structure, which also made it possible to fully describe the transition theoretically, including the precise position of the critical point and how it scaled. In this way, a long-standing gap between what can be observed in a laboratory and complicated analytical ideas is closed.
Besides the immediate results, this work offers a scalable template for exploring quantum dynamics’ limits. By showing that entanglement transitions can be reconstructed using a straightforward, linear classical decoding procedure, the team has paved the way for further research into even more intricate quantum phases and “learnability” transitions.
A public Zenodo repository has made the experiment’s information, including measurement results and circuit designs, available so that the scientific community worldwide may expand on these discoveries.
The Ohio State University, the University of Texas at Austin, and the Université de Sherbrooke collaborated on this project. It received financing from a number of organizations, including the Natural Sciences and Engineering Research Council of Canada and the Québec Ministry of Economy, Innovation, and Energy.
You can also read April McClain Delaney Reveals Federal AI & Quantum Standards
