Quantum state smoothing when Alice assumes the wrong type of monitoring by Bob
Areeya Chantasri, Kiarn T. Laverick, Howard M. Wiseman · 2025
Read Quantum state smoothing when Alice assumes the wrong type of monitoring by Bob by Areeya Chantasri, Kiarn T. Laverick, Howard M. Wiseman (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-15951.txt.
Introduction
Areeya Chantasri, Laverick, and Wiseman study quantum smoothing when one observer guesses wrong in a 2025 paper. A leaky quantum system sheds information into its surroundings, which two observers can each partly catch. Smoothing uses records from both sides of a moment to estimate the state then.
One observer must infer the other's record, which requires knowing the other's setup. The authors ask what happens when that guess is wrong. Using a glowing atom as the test system, they find accuracy swings widely by setup mix. Wrong smoothing sometimes nearly matches correct smoothing, and sometimes trails even simple filtering. Oddly, one fidelity score can favor the wrong estimate.
Catalog record
| CREATOR | Areeya Chantasri, Kiarn T. Laverick, Howard M. Wiseman |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 350 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-15951
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Quantum state smoothing when Alice assumes the wrong type of monitoring by Bob Areeya Chantasri, Kiarn T. Laverick, Howard M. Wiseman Abstract An open quantum system leaks information into its environment. In some circumstances it is possible for an observer, say Alice, to recover that information, as a classical measurement record, in a variety of different ways, using different experimental setups. The optimal way for Alice to estimate the quantum state at time t t from the record before t t is known as quantum filtering. Recently, a version of quantum smoothing, in which Alice estimates the state at time t t using her record on both sides of t t , has been developed. It requires Alice to make optimal inferences about the pre- t t record of a second observer, say Bob, who recovers whatever information Alice does not. But for Alice to make this inference, she needs to know Bob’s setup. In this paper we consider what happens if Alice is mistaken in her assumption about Bob’s setup. We show that the accuracy — as measured by the Trace-Squared-Deviation, of Alice’s estimate of the true state ( i.e. , the state conditioned on her and Bob’s pre- t t records) — depends strongly on her setup, Bob’s actual setup, and the wrongly assumed setup. Using resonance fluorescence as a model system, we show numerically that in some cases the wrong smoothing is almost as accurate as the right smoothing, but in other cases much less accurate, even being less accurate than Alice’s filtered estimate. Curiously, in some of the latter cases the fidelity of Alice’s wrong estimate with the true state is actually higher than that of her right estimate. We explain this, and other features we observe numerically, by some simple analytical arguments. 1 Introduction Many theoretical techniques applied to quantum systems have been borrowed from those successfully implemented in classical regimes. However, some have posed new puzzles, which could also lead to new discoveries. In this work, we are interested in problems of continuously monitored systems, where unknown physical states are to be estimated given partial information from observation. Two of the techniques in classical signal processing are filtering [ 1 , 2 , 3 ] and smoothing [ 4 , 5 , 6 ] , where noisy measurement signals can be ‘filtered’ (causally) or ‘smoothed’ (acausally) to estimate the underlining physical state dynamics of the system in time. The filtering technique can be readily extended to the quantum realm [ 7 , 8 , 9 , 10 ] . However, the extension of smoothing is less straight-forward, because it involves estimating a state using future information. This has led to various formalisms, including the two-state formalism [ 11 , 12 , 13 ] (generalized as the past quantum state formalism [ 14 , 15 ] ), weak values [ 16 ] and the weak-valued state [ 14 , 17 ] , smoothed psuedo-probability distributions [ 18 , 19 ] and the most likely path formalism [ 20 , 21 ] . In this paper, we are concerned with another formalism, known as quantum state smoothing [ 22 ] . (See Ref. [ 23 ] for a unified presentation and application of many of these formalisms to an open quantum system.) Quantum state smoothing applies to open quantum systems where information is missing. That is, the observer (Alice) has access to only some of the information leaking out of the system into its environment. Another observer (Bob), with access to the missing information as well as Alice’s information, can estimate the system’s properties better than Alice, allowing him to assign it a more pure state. Using quantum state smoothing, Alice can do better than using quantum state filtering, in the sense that she can assign a state that is closer to the unknown (to her, but known to Bob) state. This is a fair comparison because both quantum state smoothing and filtering can be defined as the task of optimally estimating the unknown state with the cost function being the trace-square-deviation [ 23 ] or, equivalently, the relative entropy [ 24 ] . The only difference is that quantum smoothing, like classical smoothing, uses Alice’s future record as well as her past record. However, because of the differences between classical and quantum mechanics, quantum state smoothing has some differences from the classical case, even in situations where one might assume they are the same [ 25 ] . Here, we are concerned with the fact that, contrary to classical state smoothing, the quantum version requires Alice to know how Bob is obtaining his information from the system’s environment. That is, Bob has to choose a particular measurement setup to monitor the environment, and Alice has to know what that setup is. This difference raises intriguing questions about how the quantum state smoothing power varies with different setups of the unobserved (by Alice) channels [ 26 , 27 ] . Here, by smoothing power, we mean how much improvement smoothing offers over filtering, in terms of the cost function. In fact, by choosing different setups, Bob changes his ‘unraveling’ [ 8 ] of the master equation, which can result in quantum smoothed states with quite different properties [ 25 , 26 , 28 ] . In this paper, we pose an even more dramatic question for the quantum state smoothing formalism: what would happen if Alice made a wrong assumption about what measurement setup Bob was using? How badly does this affect her ability to estimate the actual unknown state? Could smoothing even become worse than filtering? Does it depend on the characteristics of Bob’s actual unraveling in relation to the wrongly assumed one? And does it depend on the characteristics of Alice’s unraveling in relation to one or the other of these? To address these questions, after detailing the quantum state smoothing formalism in Sec. 2 , we introduce in Sec. 3 the simple system we will focus upon. This is a driven two-level atom where three measurement setups are considered: photon detection, x -homodyne detection, and y -homodyne detection, as analyzed in Ref. [ 26 ] . In Sec. 4 , we make some broad conjectures about smoothing power for all 18 combinations of Alice’s setup, Bob’s (actual) setup, and a (different) wrongly assumed setup for Bob. Numerical simulations in Sec. 5 validate these conjectures, and also allow us to dig more deeply into the questions in the preceding paragraph. We find that with a wrongly assumed unobserved (Bob’s) setup, the smoothing power (defined using the cost function) is sometimes roughly the same as that with the rightly assumed setup, but in other cases significantly worse than this, or even negative. This last case indicates that the quantum state smoothing is worse than the quantum state filtering (which is independent of Bob’s measurement), but in this case, paradoxically, the smoothed state typically has a higher fidelity with the unknown state than does the filtered state. We give some analytical explanations supplemented by numerical results for these observations in Sec. 6. Finally, we conclude in Sec. 7 . 2 Quantum State Smoothing Let us consider an open quantum system coupled to two independent baths (or