Qvine uses vine copula-inspired quantum circuit structures to achieve linear or quadratic depth scaling for loading high-dimensional distributions with high approximation quality.
hub
PennyLane: Automatic differentiation of hybrid quantum-classical computations
72 Pith papers cite this work. Polarity classification is still indexing.
72
Pith papers citing it
abstract
PennyLane is a Python 3 software framework for differentiable programming of quantum computers. The library provides a unified architecture for near-term quantum computing devices, supporting both qubit and continuous-variable paradigms. PennyLane's core feature is the ability to compute gradients of variational quantum circuits in a way that is compatible with classical techniques such as backpropagation. PennyLane thus extends the automatic differentiation algorithms common in optimization and machine learning to include quantum and hybrid computations. A plugin system makes the framework compatible with any gate-based quantum simulator or hardware. We provide plugins for hardware providers including the Xanadu Cloud, Amazon Braket, and IBM Quantum, allowing PennyLane optimizations to be run on publicly accessible quantum devices. On the classical front, PennyLane interfaces with accelerated machine learning libraries such as TensorFlow, PyTorch, JAX, and Autograd. PennyLane can be used for the optimization of variational quantum eigensolvers, quantum approximate optimization, quantum machine learning models, and many other applications.
hub tools
citation-role summary
background 1
citation-polarity summary
claims ledger
- abstract PennyLane is a Python 3 software framework for differentiable programming of quantum computers. The library provides a unified architecture for near-term quantum computing devices, supporting both qubit and continuous-variable paradigms. PennyLane's core feature is the ability to compute gradients of variational quantum circuits in a way that is compatible with classical techniques such as backpropagation. PennyLane thus extends the automatic differentiation algorithms common in optimization and machine learning to include quantum and hybrid computations. A plugin system makes the framework co
co-cited works
roles
background 1polarities
background 1representative citing papers
QLAM extends state-space models with quantum superposition in the hidden state for linear-time long-sequence modeling and reports consistent gains over RNN and transformer baselines on sequential image tasks.
Optimal FALQON optimizes per-layer δ_k and M_k via classical methods, yielding statistically significant gains in success probability and efficiency over standard FALQON on 94 non-isomorphic 3-regular graphs with 12 vertices.
Gated QKAN-FWP combines fast weight programming with quantum-inspired Kolmogorov-Arnold networks via single-qubit DARUAN activations and gated updates to deliver a 12.5k-parameter model that outperforms larger classical RNNs on long-horizon solar forecasting while running on NISQ devices.
At fixed encoding budget, serial QNN architectures suffer unbounded structural gradient starvation via rank(J) ≤ 2L+1 while parallel ones keep full Jacobian rank and better parameter efficiency when adding feature-map layers.
The hardware-compatible Brick-Circuit generator produces quantum test states with higher expressibility and entanglement than existing generators at shallower circuit depths.
QIBP adapts interval bound propagation to quantum neural networks for certified adversarial robustness via interval and affine arithmetic implementations.
Hybrid quantum walks with optimal dynamical coin operators outperform QAOA on Max-Cut and MIS by accessing a strictly larger Jordan-Lie algebra that enables faster convergence and higher accuracy.
A graph-conditioned meta-optimizer learns QAOA parameter trajectories from one problem class and transfers them to others, yielding better initializations than standard methods in an empirical study of 64 settings.
A new method for unitary synthesis on quantum hardware cuts CNOT gates by up to 36% and compiles up to 553 times faster than standard tools on square and heavy-hex lattices.
Q-SINDy adds quantum kernels to SINDy and proves that orthogonalization eliminates coefficient cannibalization bias exactly, recovering equations as accurately as classical SINDy on six tested systems.
LP2B encoding converts Lund plane jet representations into Bloch sphere qubit states, enabling a QTTN that matches classical LundNet performance on polarization tagging and W/top tagging with three orders of magnitude fewer parameters and improved low-data regime results.
Hybrid quantum PINN for hydrology reports 3x faster convergence and 44% fewer parameters than classical PINN on Sri Lankan flood data while using physics constraints for uncertainty quantification.
A quantum residual neural network implements skip connections via fixed linear mixing of identity and variational circuits, achieving 99% binary and 80% multi-class image classification accuracy with 10x fewer gates than standard variational models while mitigating barren plateaus and showing Advers
A formal audit of 45 quantum simulators identifies 547 vulnerabilities across memory corruption, resource exhaustion, code injection, and a new QASM injection class, with all patterns verified by Z3 SAT proofs.
HQ-LP-FNO replaces part of the spectral channel mixing in a 3D FNO with a mode-shared VQC, reducing parameters by 15.6% and phase-fraction MAE by 26% on laser-processing surrogates while remaining stable under calibrated noise.
CUTS-GPR performs numerically exact Gaussian process regression with near-linear scaling in training points N and low-order polynomial scaling in dimensions D by exploiting additive kernels on incomplete grids.
Entanglement improves classification accuracy in distributed quantum ML tasks across datasets, but excessive amounts degrade performance by reducing effective parameter dimension.
Hybrid agent with variational quantum circuits for feature extraction in hierarchical RL outperforms classical baselines with 66% parameter savings, but quantum value estimation degrades results.
QTL unifies expectation-value minimization with CVaR and Gibbs heuristics under one tunable operator, amplifying gradients in structured cases while preserving global minima and shifting the bottleneck to measurement variance.
GQPINNs add symmetry awareness to quantum PINNs via equivariant circuits, yielding lower mean absolute error and fewer parameters than standard QPINNs on linear and nonlinear PDE benchmarks.
A single-qubit quantum reinforcement learning agent solves CartPole faster than classical networks and quantifies shot-count versus control-frequency requirements for real-time closed-loop control on NISQ hardware, including direct electronics programming to reduce latency.
Hardware-efficient gates are universal for state preparation in particle-number and symmetry-constrained subspaces because commutators generate Pauli Z projectors that span the full so(w) and su(w) algebras.
A quantum autoencoder purifies adversarial perturbations for quantum classifiers and supplies a confidence score for unrecoverable inputs, claiming up to 68% accuracy gains over prior defenses without adversarial training.
citing papers explorer
-
Qvine: Vine Structured Quantum Circuits for Loading High Dimensional Distributions
Qvine uses vine copula-inspired quantum circuit structures to achieve linear or quadratic depth scaling for loading high-dimensional distributions with high approximation quality.
-
QLAM: A Quantum Long-Attention Memory Approach to Long-Sequence Token Modeling
QLAM extends state-space models with quantum superposition in the hidden state for linear-time long-sequence modeling and reports consistent gains over RNN and transformer baselines on sequential image tasks.
-
Optimal FALQON for Quantum Approximate Optimization via Layer-wise Parameter Tuning
Optimal FALQON optimizes per-layer δ_k and M_k via classical methods, yielding statistically significant gains in success probability and efficiency over standard FALQON on 94 non-isomorphic 3-regular graphs with 12 vertices.
-
Gated QKAN-FWP: Scalable Quantum-inspired Sequence Learning
Gated QKAN-FWP combines fast weight programming with quantum-inspired Kolmogorov-Arnold networks via single-qubit DARUAN activations and gated updates to deliver a 12.5k-parameter model that outperforms larger classical RNNs on long-horizon solar forecasting while running on NISQ devices.
-
Architecture Shape Governs QNN Trainability: Jacobian Null Space Growth and Parameter Efficiency
At fixed encoding budget, serial QNN architectures suffer unbounded structural gradient starvation via rank(J) ≤ 2L+1 while parallel ones keep full Jacobian rank and better parameter efficiency when adding feature-map layers.
-
Randomized and Diverse Input State Generation for Quantum Program Testing
The hardware-compatible Brick-Circuit generator produces quantum test states with higher expressibility and entanglement than existing generators at shallower circuit depths.
-
Quantum Interval Bound Propagation for Certified Training of Quantum Neural Networks
QIBP adapts interval bound propagation to quantum neural networks for certified adversarial robustness via interval and affine arithmetic implementations.
-
Beyond Single Trajectories: Optimal Control and Jordan-Lie Algebra in Hybrid Quantum Walks for Combinatorial Optimization
Hybrid quantum walks with optimal dynamical coin operators outperform QAOA on Max-Cut and MIS by accessing a strictly larger Jordan-Lie algebra that enables faster convergence and higher accuracy.
-
Graph-Conditioned Meta-Optimizer for QAOA Parameter Generation on Multiple Problem Classes
A graph-conditioned meta-optimizer learns QAOA parameter trajectories from one problem class and transfers them to others, yielding better initializations than standard methods in an empirical study of 64 settings.
-
Architecture-aware Unitary Synthesis
A new method for unitary synthesis on quantum hardware cuts CNOT gates by up to 36% and compiles up to 553 times faster than standard tools on square and heavy-hex lattices.
-
Q-SINDy: Quantum-Kernel Sparse Identification of Nonlinear Dynamics with Provable Coefficient Debiasing
Q-SINDy adds quantum kernels to SINDy and proves that orthogonalization eliminates coefficient cannibalization bias exactly, recovering equations as accurately as classical SINDy on six tested systems.
-
Lund Plane to Bloch (LP2B) Encoding for Object and Polarization Tagging with Quantum Jet Substructure
LP2B encoding converts Lund plane jet representations into Bloch sphere qubit states, enabling a QTTN that matches classical LundNet performance on polarization tagging and W/top tagging with three orders of magnitude fewer parameters and improved low-data regime results.
-
Variational Quantum Physics-Informed Neural Networks for Hydrological PDE-Constrained Learning with Inherent Uncertainty Quantification
Hybrid quantum PINN for hydrology reports 3x faster convergence and 44% fewer parameters than classical PINN on Sri Lankan flood data while using physics constraints for uncertainty quantification.
-
A hardware efficient quantum residual neural network without post-selection
A quantum residual neural network implements skip connections via fixed linear mixing of identity and variational circuits, achieving 99% binary and 80% multi-class image classification accuracy with 10x fewer gates than standard variational models while mitigating barren plateaus and showing Advers
-
Broken Quantum: A Systematic Formal Verification Study of Security Vulnerabilities Across the Open-Source Quantum Computing Simulator Ecosystem
A formal audit of 45 quantum simulators identifies 547 vulnerabilities across memory corruption, resource exhaustion, code injection, and a new QASM injection class, with all patterns verified by Z3 SAT proofs.
-
Hybrid Fourier Neural Operator for Surrogate Modeling of Laser Processing with a Quantum-Circuit Mixer
HQ-LP-FNO replaces part of the spectral channel mixing in a 3D FNO with a mode-shared VQC, reducing parameters by 15.6% and phase-fraction MAE by 26% on laser-processing surrogates while remaining stable under calibrated noise.
-
Don't Get Your Kroneckers in a Twist: Gaussian Processes on High-Dimensional Incomplete Grids
CUTS-GPR performs numerically exact Gaussian process regression with near-linear scaling in training points N and low-order polynomial scaling in dimensions D by exploiting additive kernels on incomplete grids.
-
The power of entanglement in distributed quantum machine learning
Entanglement improves classification accuracy in distributed quantum ML tasks across datasets, but excessive amounts degrade performance by reducing effective parameter dimension.
-
Quantum Hierarchical Reinforcement Learning via Variational Quantum Circuits
Hybrid agent with variational quantum circuits for feature extraction in hierarchical RL outperforms classical baselines with 66% parameter savings, but quantum value estimation degrades results.
-
Quantum Tilted Loss in Variational Optimization: Theory and Applications
QTL unifies expectation-value minimization with CVaR and Gibbs heuristics under one tunable operator, amplifying gradients in structured cases while preserving global minima and shifting the bottleneck to measurement variance.
-
Geometric Quantum Physics Informed Neural Network
GQPINNs add symmetry awareness to quantum PINNs via equivariant circuits, yielding lower mean absolute error and fewer parameters than standard QPINNs on linear and nonlinear PDE benchmarks.
-
Towards Real-time Control of a CartPole System on a Quantum Computer
A single-qubit quantum reinforcement learning agent solves CartPole faster than classical networks and quantifies shot-count versus control-frequency requirements for real-time closed-loop control on NISQ hardware, including direct electronics programming to reduce latency.
-
Universality of Quantum Gates in Particle and Symmetry Constrained Subspaces
Hardware-efficient gates are universal for state preparation in particle-number and symmetry-constrained subspaces because commutators generate Pauli Z projectors that span the full so(w) and su(w) algebras.
-
Defending Quantum Classifiers against Adversarial Perturbations through Quantum Autoencoders
A quantum autoencoder purifies adversarial perturbations for quantum classifiers and supplies a confidence score for unrecoverable inputs, claiming up to 68% accuracy gains over prior defenses without adversarial training.
-
Controlled Steering-Based State Preparation for Adversarial-Robust Quantum Machine Learning
A passive steering method for quantum state preparation improves adversarial accuracy in QML models by up to 40% across tested cases.
-
QuanForge: A Mutation Testing Framework for Quantum Neural Networks
QuanForge introduces statistical mutation killing and nine post-training mutation operators for QNNs to distinguish test suites and localize vulnerable circuit regions.
-
Assessing System Capabilities and Bottlenecks of an Early Fault-Tolerant Bicycle Architecture
Syn@fac optimization reduces estimated circuit failure probability by a factor of 9 on average across non-Clifford benchmarks for bivariate bicycle code modular FTQC architectures, with additional gains from transvection deferral and Clifford insertion.
-
Block-encodings as programming abstractions: The Eclipse Qrisp BlockEncoding Interface
The Eclipse Qrisp BlockEncoding interface provides high-level programming abstractions for block-encodings, enabling easier implementation of quantum algorithms such as QSVT, matrix inversion, and Hamiltonian simulation.
-
Hierarchical Progressive Pauli Noise Modeling with Residual Compensation for Multi-Qubit Quantum Circuits
HPO framework reduces multi-qubit Pauli noise characterization complexity from O(4^N) to O(N·4^w) with 96.3% parameter compression on 5 qubits and raises 10-qubit HHL fidelity from 0.7431 to 0.9381.
-
Double Descent in Quantum Kernel Ridge Regression
Quantum kernel ridge regression shows double descent in test risk, with the interpolation peak suppressible by regularization, via random matrix theory asymptotics in the high-dimensional limit.
-
Observable-Guided Generator Selection for Improving Trainability in Quantum Machine Learning with a $ \mathfrak{g} $-Purity Interpretation under Restricted Settings
Observable-guided selection of anti-commuting Pauli generators improves training speed in 5-qubit QML circuits and admits a g-purity interpretation under restricted algebraic assumptions.
-
QLLVM: A Scalable Quantum-Classical Co-Compilation Framework based on LLVM
QLLVM delivers an LLVM-based end-to-end co-compiler that unifies classical HPC and quantum programs into one executable, with a three-stage quantum path via MLIR and QIR that reduces circuit depth and gate counts on MQTBench versus prior compilers.
-
Learning to Concatenate Quantum Codes
A machine-learning approach adaptively chooses quantum code sequences for concatenation to achieve target logical error rates with far fewer qubits than standard methods for structured noise.
-
Parameter-efficient Quantum Multi-task Learning
QMTL uses shared VQC encoding plus task-specific quantum ansatz heads to achieve linear parameter scaling with the number of tasks while matching or exceeding classical multi-task baselines on three benchmarks.
-
Quantum-Enhanced Single-Parameter Phase Estimation with Adaptive NOON States
Optimizing eight circuit parameters via gradient descent on classical Fisher information yields up to 1598% CFI gains and 133x more useful events per pulse for N=5 NOON states, pushing performance closer to the Heisenberg limit than the Afek et al. 2010 setup.
-
Adaptive H-EFT-VA: A Provably Safe Trajectory Through the Trainability-Expressibility Landscape of Variational Quantum Algorithms
Adaptive H-EFT-VA maintains gradient variance Omega(1/poly(N)) during safe Hilbert space expansion, doubling fidelity over static H-EFT-VA on benchmarks up to 14 qubits.
-
Quantum Patches: Enhancing Robustness of Quantum Machine Learning Models
Random quantum circuits used as adversarial training data reduce successful attack rates on QML models for CIFAR-10 from 89.8% to 68.45% and for CINIC-10 from 94.23% to 78.68%.
-
Shot-Based Quantum Encoding: A Data-Loading Paradigm for Quantum Neural Networks
SBQE encodes data via learnable shot distributions over initial states to form mixed quantum representations, achieving 89.1% accuracy on Semeion and 80.95% on Fashion MNIST without encoding gates.
-
Molecular Excited States using Quantum Subspace Methods: Accuracy, Resource Reduction, and Error-Mitigated Hardware Implementation of q-sc-EOM
Optimized q-sc-EOM on quantum hardware yields accurate excited-state energies for challenging molecular bond-breaking cases after reducing measurement scaling to O(N^5) and applying readout and symmetry error mitigation.
-
Eliminating Vendor Lock-In in Quantum Machine Learning via Framework-Agnostic Neural Networks
A new QNN architecture with unified graph, HAL, and ONNX pipeline enables cross-framework and cross-hardware QML with training time within 8% of native implementations and identical accuracy on Iris, Wine, and MNIST-4 tasks.
-
Wavelet Variance Equipartition as a Threshold for World-Model Quality and Quantum Kernel TN-Simulability
Wavelet scaling α = 1/2 separates classically simulable area-law from volume-law phases for quantum kernels in world-model latents, with empirical VideoMAE latents and a Θ(d^{-2}) variance bound implying simulation hardness and quadratic measurement costs.
-
Classic and Quantum Task-Based Intelligent Runtime for QIRs Running on Multiple QPUs
A hybrid runtime system merges classical task scheduling with quantum execution to enable concurrent QIR program dispatch to multiple QPUs and simulators, demonstrated through circuit cutting on 4- and 20-qubit examples.
-
The Quantum Hamiltonian Analysis Toolkit: Lowering the Barrier to Quantum Computing with Hamiltonians
QHAT is a user-friendly software toolkit for Hamiltonian generation, analysis, and fault-tolerant quantum simulation driven by error tolerances rather than algorithmic parameters.
-
Measuring Accuracy and Energy-to-Solution of Quantum Fine-Tuning of Foundational AI Models
Trapped-ion quantum fine-tuning of AI models shows linear energy scaling and 24% better classification error than classical logistic regression or SVM baselines, with a projected energy break-even at 34 qubits.
-
On the importance of hyperparameters in initializing parameterized quantum circuits
An evolutionary algorithm optimizes initialization hyperparameters for quantum circuits, leading to faster convergence without worsening barren plateaus.
-
Quantum-HPC Software Stacks and the openQSE Reference Architecture: A Survey
A survey of nine QHPC stacks identifies common patterns and proposes the openQSE reference architecture to unify interfaces for interoperability in quantum-HPC environments.
-
Physics-Informed Neural Networks for Maximizing Quantum Fisher Information in Time-Dependent Many-Body Systems
PINNs combined with Magnus expansion learn scheduling functions and adiabatic gauge potentials that yield higher normalized QFI than Euler-Lagrange baselines in nearest-neighbor, dipolar, and trapped-ion spin models up to six qubits.
-
Warm-Start Quantum Approximate Optimization Algorithm for QAM MIMO Data Detection
WSLR-QAOA combines a low-rank semidefinite warm-start with linear-ramp QAOA to achieve symbol error rates close to maximum-likelihood detection for high-order QAM MIMO while running faster than standard QAOA and showing usable performance on IBM quantum hardware.
-
Continuous-time quantum-walk centrality for protein residue interaction networks
Continuous-time quantum walks on protein residue networks yield a centrality measure that agrees with eigenvector centrality, incorporates quantum interference, recovers functional residues, and runs on IBM quantum hardware.
-
How Embeddings Shape Graph Neural Networks: Classical vs Quantum-Oriented Node Representations
Quantum-oriented embeddings deliver consistent gains on structure-driven graph datasets while classical baselines perform adequately on attribute-limited social graphs, under identical training pipelines across five TU datasets and binned QM9.