SchrodingeRNN: Generative Modeling of Raw Audio as a Continuously Observed Quantum State

被引:0
|
作者
Uranga, Benat Mencia [1 ]
Lamacraft, Austen [1 ]
机构
[1] Univ Cambridge, TCM Grp, Cavendish Lab, JJ Thomson Ave, Cambridge CB3 0HE, England
来源
MATHEMATICAL AND SCIENTIFIC MACHINE LEARNING, VOL 107 | 2020年 / 107卷
基金
英国工程与自然科学研究理事会;
关键词
Machine Learning; Generative Models; Quantum Physics; Matrix Product States;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We introduce SchrodingeRNN, a quantum-inspired generative model for raw audio. Audio data is wave-like and is sampled from a continuous signal. Although generative modeling of raw audio has made great strides lately, relational inductive biases relevant to these two characteristics are mostly absent from models explored to date. Quantum Mechanics is a natural source of probabilistic models of wave behavior. Our model takes the form of a stochastic Schrodinger equation describing the continuous time measurement of a quantum system, and is equivalent to the continuous Matrix Product State (cMPS) representation of wavefunctions in one dimensional many-body systems. This constitutes a deep autoregressive architecture in which the system's state is a latent representation of the past observations. We test our model on synthetic data sets of stationary and non-stationary signals. This is the first time cMPS are used in machine learning.
引用
收藏
页码:74 / 106
页数:33
相关论文
共 50 条
  • [21] Quantum State Tomography with Conditional Generative Adversarial Networks
    Ahmed, Shahnawaz
    Sanchez Munoz, Carlos
    Nori, Franco
    Kockum, Anton Frisk
    PHYSICAL REVIEW LETTERS, 2021, 127 (14)
  • [22] SCATTERING-THEORY FOR THE BELAVKIN EQUATION DESCRIBING QUANTUM PARTICLE WITH CONTINUOUSLY OBSERVED COORDINATE
    KOLOKOLTSOV, VN
    JOURNAL OF MATHEMATICAL PHYSICS, 1995, 36 (06) : 2741 - 2760
  • [23] A generative modeling approach for benchmarking and training shallow quantum circuits
    Benedetti, Marcello
    Garcia-Pintos, Delfina
    Perdomo, Oscar
    Leyton-Ortega, Vicente
    Nam, Yunseong
    Perdomo-Ortiz, Alejandro
    NPJ QUANTUM INFORMATION, 2019, 5 (1)
  • [24] Variational measurement-based quantum computation for generative modeling
    Majumder, Arunava
    Krumm, Marius
    Radkohl, Tina
    Fiderer, Lukas J.
    Nautrup, Hendrik Poulsen
    Jerbi, Sofiene
    Briegel, Hans J.
    PHYSICAL REVIEW A, 2024, 110 (06)
  • [25] A generative modeling approach for benchmarking and training shallow quantum circuits
    Marcello Benedetti
    Delfina Garcia-Pintos
    Oscar Perdomo
    Vicente Leyton-Ortega
    Yunseong Nam
    Alejandro Perdomo-Ortiz
    npj Quantum Information, 5
  • [26] Multiqubit state learning with entangling quantum generative adversarial networks
    Rasmussen, S. E.
    Zinner, N. T.
    PHYSICAL REVIEW A, 2022, 106 (03)
  • [27] QuGAN: A Quantum State Fidelity based Generative Adversarial Network
    Stein, Samuel A.
    Baheri, Betis
    Chen, Daniel
    Mao, Ying
    Guan, Qiang
    Li, Ang
    Fang, Bo
    Xu, Shuai
    2021 IEEE INTERNATIONAL CONFERENCE ON QUANTUM COMPUTING AND ENGINEERING (QCE 2021) / QUANTUM WEEK 2021, 2021, : 71 - 81
  • [28] Correlations of the Time Dependent Signal and the State of a Continuously Monitored Quantum System
    Foroozani, N.
    Naghiloo, M.
    Tan, D.
    Molmer, K.
    Murch, K. W.
    PHYSICAL REVIEW LETTERS, 2016, 116 (11)
  • [29] A novel ground state and excitation observed in a quantum magnet
    Katsumata, K
    Matsuda, M
    JOURNAL OF THE PHYSICAL SOCIETY OF JAPAN, 2000, 69 : 303 - 310
  • [30] Quantum-probabilistic Hamiltonian learning for generative modeling and anomaly detection
    Araz, Jack Y.
    Spannowsky, Michael
    PHYSICAL REVIEW A, 2023, 108 (06)