Realizing a deep reinforcement learning agent for real-time quantum feedback

Reuer, Kevin; Landgraf, Jonas; Fösel, Thomas; O’Sullivan, James; Beltrán, Liberto; Akin, Abdulkadir; Norris, Graham J.; Remm, Ants; Kerschbaum, Michael; Besse, Jean-Claude; Marquardt, Florian; Wallraff, Andreas; Eichler, Christopher

doi:10.1038/s41467-023-42901-3

Cited by 9 publications

(3 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With an L of 12 µm, the capacitance sensitivity of the interdigital capacitor decreases to about 2.1 fF/µm, and the overall device size is 250 µm×317 µm, 2.3 times larger than the previous pattern. Another strategy involves the use of an enclosing capacitor [23][24][25] as depicted in Fig. 3.…”

Section: Sensitivity Analysis Of Coupling Capacitancementioning

confidence: 99%

Optimize Purcell filter design for reducing influence of fabrication variation

Cai 蔡,

Zhou 周,

Yu 于

et al. 2024

Chinese Phys. B

View full text Add to dashboard Cite

To protect superconducting qubits and enable rapid readout, optimally designed Purcell filters are essential. To suppress the off-resonant driving of untargeted readout resonators, individual Purcell filters were used for each readout resonator. However, achieving consistent frequency between a readout resonator and a Purcell filter is a significant challenge. A systematic computational analysis was conducted to investigate how fabrication variation impacts filter performance, focusing on the coupling capacitor structure and coplanar waveguide (CPW) transmission line specifications. The results indicate that the T-type enclosing capacitor (EC), which exhibits lower structural sensitivity, is more advantageous for achieving target capacitance compared to the C-type EC and the interdigital capacitor (IDC). By utilizing a large-sized CPW with the T-type EC structure, fluctuations in the effective coupling strength can be reduced to 10% given typical micro-nanofabrication variances. The numerical simulations presented in this work minimize the influence of fabrication deviations, thereby significantly improving the reliability of Purcell filter designs.

show abstract

Section: Sensitivity Analysis Of Coupling Capacitancementioning

confidence: 99%

Optimize Purcell filter design for reducing influence of fabrication variation

Cai 蔡,

Zhou 周,

Yu 于

et al. 2024

Chinese Phys. B

View full text Add to dashboard Cite

show abstract

“…Artificial intelligence (AI) in the form of reinforcement learning has also been widely explored as a viable method to reveal new physics and improve established numerical and experimental methods. For example, AI models were used to simulate the design of photonic experiments with entangled states [28], discover quantum error correction strategies [29,30] and generate elementary gate sequences from a quantum algorithm, therefore making it suitable for hardware implementation [31].…”

Section: Introductionmentioning

confidence: 99%

Design of Nanoscale Quantum Interconnects Aided by Conditional Generative Adversarial Networks

Preda,

Pantis-Simut,

Marciu

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

Interconnecting nanodevices with the aim of assembling quantum computing architectures is one of the current outstanding challenges. At the nanoscale, the quantum interconnects become comparable in complexity with the active devices and should be treated on equal footing. In addition, they can play an active role in the switching properties. Here, we investigate the charge localization in neuromorphic bi-dimensional systems, which serve as quantum interconnects (QIs) between quantum dot registers. We define a device structure where, by manipulating the charging of a floating gate array, one defines the QI potential map, which can host a few interacting electrons. The ground state charge density may be extracted by measuring the tunneling current perpendicular to the device surface, yielding a convoluted image of the electron distribution. Using image-to-image translation methods, we achieve the mapping of the charge density from the confinement potential, as well as by deconvoluting the tunneling current map, which can be obtained by a direct measurement. Thus, we provide a proof-of-concept for a reconfigurable device, which can be used to design quantum many-electron devices.

show abstract

“…AI pipelines excel at this task proposing new experimental setups based on iterative feedback [144], with most applications focusing on quantum optics [22,[145][146][147]. Furthermore, they also prove invaluable tools to control the experimental parameters both in offline [148][149][150] and online [151][152][153], facilitating the development of advanced protocols such as rapid cooling schemes [154,155]. Given the substantial data output of quantum experiments, ML algorithms emerge among the best tools for the result analysis.…”

Section: Machine Learning For Quantum Many-body Physicsmentioning

confidence: 99%

A machine learning ride in the physics theme park: from quantum to biophysics

Requena Pozo

View full text Add to dashboard Cite

(English) The integration of artificial intelligence into research is propelling progress and discoveries across the entire scientific landscape. Artificial intelligence tools boost the development of novel scientific insights and theories by processing extensive data sets, guiding exploration and hypothesis formation, enhancing experimental setups, and even enabling autonomous discovery. In this thesis, we harness the power of machine learning, a sub-field of artificial intelligence, to study non-deterministic systems, which are amongst the hardest to characterize. On one hand, we address problems inherent to the study of quantum systems and the development of quantum technologies. Quantum physics presents formidable challenges due to the associated exponential complexity with the size of the system at hand, as well as its intrinsic stochastic nature and the presence of intricate correlations between its components. We employ reinforcement learning, a machine learning technique that excels at dealing with vast hypothesis spaces, to address some of these challenges. Notably, reinforcement learning has demonstrated super-human performance in multiple complex games like Go, which present similar characteristics to the problems encountered in the study of quantum physics. We use it to systematically simplify complex common problems in condensed matter and quantum information processing tasks, as well as to implement robust calibration schemes for quantum computers. On the other hand, we focus on the characterization of complex stochastic processes, such as diffusion. Understanding diffusion processes is crucial to unravel the complex underlying physical and biological mechanisms governing them. This involves extracting meaningful parameters from the analysis of stochastic trajectories described by tracked particles. However, accurately capturing and analyzing the trajectories presents multiple challenges, stemming from the combination of their random nature, complex dynamics, and experimental drawbacks, such as noise. We develop machine learning algorithms to accurately extract such parameters, even when they vary with time, and demonstrate their applicability in experimental scenarios. Furthermore, we apply similar techniques to study the diffusion of internet users browsing an e-commerce website, predicting their likelihood to make a purchase before closing the session. (Català) La integració de la intel·ligència artificial a la recerca accelera el progrés cap a nous descobriments en tot l’'àmbit científic. Les eines d’'intel·ligència artificial contribueixen al desenvolupament de noves teories i coneixements processant grans quantitats de dades, guiant l’'exploració i la formulació d'’hipòtesis, millorant els experiments i, fins i tot, fent possible descobriments automàtics. En aquesta tesi, aprofitem el poder de l’'aprenentatge automàtic (“"machine learning”"), un camp de la intel·ligència artificial, per estudiar sistemes no-deterministes, que es troben entre els més difícils de caracteritzar. Per una banda, tractem problemes inherents a l’'estudi de sistemes quàntics i del desenvolupament de noves tecnologies quàntiques. La física quàntica planteja reptes formidables derivats de la complexitat exponencial amb la mida del sistema considerat, en combinació amb una naturalesa intrínsicament estocàstica i la presència de correlacions complexes entre elements del sistema. Per tractar alguns d’'aquests reptes, fem servir aprenentatge de reforç (“"reinforcement leraning”"), una tècnica de l'aprenentatge automàtic capaç d’'explorar grans espais d'’hipòtesis. Per exemple, empleant aquestes tècniques, s'’ha aconseguit superar als millors jugadors del món en jocs complexes com el Go, que presenten característiques similars a problemes emergents en l’'estudi de la quàntica. En el nostre cas, desenvolupem mètodes per simplificar problemes complexes comuns en els camps de la matèria condensada i de la informació quàntica de forma sistemàtica, i disenyem protocols robustos de cal·libració d’'ordinadors quàntics. Per l’altra banda, ens dediquem a la caracterització de processos estocàstics complexes, com és la difusió. Entendre els processos de difusió és essencial per descobrir els mecanismes físics i biològics que els governen, el que comporta l’'anàlisi de trajectòries estocàstiques descrites per partícules per tal d’extruere’n paràmetres significatius del sistema. Aquest anàlisi, però, presenta grans reptes des de l’'adquisició de les trajectòries fins al seu estudi posterior que provenen, principalment, de la combinació de la seva naturales aleatòria, amb la presència de dinàmiques complexes i altres inconvenients experimentals, com ara el soroll. Utilitzant tècniques d’aprenentatge automàtic, desenvolupem algoritmes per analitzar aquestes trajectòries i extreure’n els paràmetres d’interès acuradament, fins i tot quan aquests canvien amb el temps. Després, utilitzem aquests algoritmes per estudiar diferents experiments en sistemes biològics i, també, per estudiar les trajectòries descrites per usuaris navegant una pàgina de comerç online. En aquest últim cas, en comptes d'extreure paràmetres físics, inferim si l'usuari farà una compra abans de tancar la sessió.

show abstract

Realizing a deep reinforcement learning agent for real-time quantum feedback

Cited by 9 publications

References 52 publications

Optimize Purcell filter design for reducing influence of fabrication variation

Optimize Purcell filter design for reducing influence of fabrication variation

Design of Nanoscale Quantum Interconnects Aided by Conditional Generative Adversarial Networks

A machine learning ride in the physics theme park: from quantum to biophysics

Contact Info

Product

Resources

About