• Corpus ID: 253237692

Quantum Variational Rewinding for Time Series Anomaly Detection

  title={Quantum Variational Rewinding for Time Series Anomaly Detection},
  author={Jack S Baker and Haim Horowitz and Santosh Kumar Radha and Stenio F. L. Fernandes and Colin Jones and Noor Fadiya Mohd Noor Ishak Hashim Mohd Salmi Noorani and V. Anatolievich Skavysh and Philippe Lamontangne and Barry C. Sanders},
Electron dynamics, financial markets and nuclear fission reactors, though seemingly unrelated, all produce observable characteristics evolving with time. Within this broad scope, departures from normal temporal behavior range from academically interesting to potentially catastrophic. New algorithms for time series anomaly detection (TAD) are therefore certainly in demand. With the advent of newly accessible quantum processing units (QPUs), exploring a quantum approach to TAD is now relevant and… 

Figures and Tables from this paper



Quantum machine learning for quantum anomaly detection

It is shown that kernel principal component analysis and one-class support vector machine can be performed using resources logarithmic in the dimensionality of quantum states, which makes these algorithms potentially applicable to big quantum data applications.

Hybrid classical-quantum autoencoder for anomaly detection

It is shown that the addition of the PQC to the autoencoder bottleneck leads to a performance enhancement in terms of precision, recall, and F1 score.

Variational quantum anomaly detection: Unsupervised mapping of phase diagrams on a physical quantum computer

Variational quantum anomaly detection is proposed, an unsupervised quantum machine learning algorithm to analyze quantum data from quantum simulation to extract the phase diagram of a system with no prior physical knowledge and can be performed end-to-end on the same quantum device where the system is simulated on.

Unsupervised Anomaly Detection in Energy Time Series Data Using Variational Recurrent Autoencoders with Attention

  • João PereiraM. Silveira
  • Computer Science
    2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA)
  • 2018
This paper proposes a generic, unsupervised and scalable framework for anomaly detection in time series data, based on a variational recurrent autoencoder, and introduces attention in the model, by means of a Variational self-attention mechanism (VSAM), to improve the performance of the encoding-decoding process.

Information-theoretic bounds on quantum advantage in machine learning

It is proven that for any input distribution D(x), a classical ML model can provide accurate predictions on average by accessing E a number of times comparable to the optimal quantum ML model, and it is proved that the exponential quantum advantage is possible.

Dynamical simulation via quantum machine learning with provable generalization

A framework for using QML methods to simulate quantum dynamics on near-term quantum hardware and uses generalization bounds, which bound the error a machine learning model makes on unseen data, to rigorously analyze the training data requirements of an algorithm within this framework.

Noise-induced barren plateaus in variational quantum algorithms

This work rigorously proves a serious limitation for noisy VQAs, in that the noise causes the training landscape to have a barren plateau, and proves that the gradient vanishes exponentially in the number of qubits n if the depth of the ansatz grows linearly with n.

Out-of-distribution generalization for learning quantum dynamics

Generalization bounds are a critical tool to assess the training data requirements of Quantum Machine Learning (QML). Recent work has established guarantees for in-distribution generalization of

Dynamical decoupling for superconducting qubits: a performance survey

Dynamical Decoupling (DD) is perhaps the simplest and least resource-intensive error suppression strategy for improving quantum computer performance. Here we report on a large-scale survey of the

Anomaly detection with variational quantum generative adversarial networks

This model replaces the generator of WGANs with a hybrid quantum–classical neural net and leaves the classical discriminative model unchanged, which means high-dimensional classical data only enters the classical model and need not be prepared in a quantum circuit.