Research at The University of Hong Kong and HK Institute of Quantum Science & Technology classifies computation within quantum processors. The study concentrates on energy-counting thermal-idle instruments featuring key stabiliser control together with independent local Markov baths. Stabiliser errors build up linearly with time at a fixed thermal exposure.

Thermal Noise Management Defines Stabiliser Processor Computational Power

Scientists at the University of Hong Kong and HK Institute of Quantum Science & Technology have shown retaining information about energy exchange during idle periods can unlock universal quantum computation; previously, erasing this record limited processors to classical simulation. A stabilizer processor’s ability to perform complex calculations now hinges on its controller’s memory management of thermal exposure events, demanding careful consideration of these parameters for optimal performance.

The team demonstrated how selectively preserving or discarding records enables distillable ancillas for fault tolerance, even when averaged channels appear classically simulable. Modest thermal noise can be harnessed for quantum computation if its history is carefully managed. This capability depends on the interplay between relaxation time (T₁), coherence time (T₂) and excited-state population (p_e).

Amplitude damping, a common source of error in qubits, generates useful nonstabilizer states through clever postselection techniques utilising entangled pairs known as Bell pairs. Furthermore, the research established a connection between ‘magic generation’, quantifying computational power beyond standard stabilizer circuits, and distillability with temperature considerations.

Branchwise nonnegative stabilizer decompositions provide an explicit efficient classical sampler for adaptive circuits and their full time-resolved exchange record if every location lies at or below this threshold. A suitable idle duration and no-exchange conditioning supply distillable ancillas enabling universal quantum computation with polynomial overhead when exceeding this boundary at even one repeatedly accessible location. Erasing the record after sufficiently long exposure makes averaged channels stabilizer measure-and-prepare at finite temperatures on the resource side. However, coherence is lost in doing so.

Thermal noise exploitation defines limits for strong quantum computation

These findings offer tantalising prospects for building more durable quantum processors but depend upon achieving ‘ideal stabiliser control’ and assuming perfectly isolated local environments, conditions notoriously difficult to maintain in practice. While providing an exact boundary defining when thermal noise becomes a computational asset rather than a hindrance, translating these insights into tangible hardware presents significant hurdles. Acknowledging that maintaining perfect isolation remains an engineering challenge does not diminish this theoretical advance as it pinpoints exactly what levels of environmental disturbance can be constructively harnessed instead of suppressed.

This thorough understanding allows prioritisation of specific control mechanisms and error mitigation strategies with greater accuracy, accelerating progress towards fault-tolerant computation. The team’s work demonstrates that a quantum processor’s computational power isn’t solely determined by minimising errors but also by how information about thermal disturbances is managed. Retaining or discarding data regarding energy exchange during idle periods fundamentally alters performance; consequently, processors are classified either classically simulable or capable of universal quantum computation based on record retention strategy.

This means that how thermal noise is handled impacts computational capability, with retaining information about energy exchange potentially enabling more powerful processing. The authors found that processors are classified according to their ability to process data depending on which environmental records its controller retains. They suggest further work will focus on understanding these parameters in real systems.

👉 More information
🗞 Environmental records unlock universal quantum computation from thermal decoherence
✍️ Chenfeng Cao and Qi Zhao
🧠 ArXiv: https://arxiv.org/abs/2609.16460

More like this