arXiv:2410.23018 · Physical review. E

Optimizing temperature distributions for training neural quantum states using parallel tempering

Conor Smith, Quinn Campbell, Tameem Albash

Published May 22, 2025• 0 citations

Abstract

Parametrized artificial neural networks (ANNs) can be very expressive ansatzes for variational algorithms, reaching state-of-the-art energies on many quantum many-body Hamiltonians. Nevertheless, the training of the ANN can be slow and stymied by the presence of local minima in the parameter landscape. One approach to mitigate this issue is to use parallel tempering methods, and in this work, we focus on the role played by the temperature distribution of the parallel tempering replicas. Using an adaptive method that adjusts the temperatures in order to equate the exchange probability between neighboring replicas, we show that this temperature optimization can significantly increase the success rate of the variational algorithm with negligible computational cost by eliminating bottlenecks in the replicas' random walk. We demonstrate this using two different neural networks, a restricted Boltzmann machine and a feedforward network, which we use to study a toy problem based on a permutation invariant Hamiltonian with a pernicious local minimum and the J_{1}-J_{2} model on a rectangular lattice.

DOI: 10.1103/physreve.111.055306

Related papers

Quantum-inspired tempering for ground state approximation using artificial neural networks

Tameem Albash, Conor Smith, Quinn Campbell +1 more

A large body of work has demonstrated that parameterized artificial neural networks (ANNs) can efficiently describe ground states of numerous interesting quantum many-body Hamiltonians. However, the standard variational algorithms used to update or train the ANN parameters can get trapped in local minima, especially for frustrated systems and even if the representation is sufficiently expressive. We propose a parallel tempering method that facilitates escape from such local minima. This methods involves training multiple ANNs independently, with each simulation governed by a Hamiltonian with a different "driver" strength, in analogy to quantum parallel tempering, and it incorporates an update step into the training that allows for the exchange of neighboring ANN configurations. We study instances from two classes of Hamiltonians to demonstrate the utility of our approach using Restricted Boltzmann Machines as our parameterized ANN. The first instance is based on a permutation-invariant Hamiltonian whose landscape stymies the standard training algorithm by drawing it increasingly to a false local minimum. The second instance is four hydrogen atoms arranged in a rectangle, which is an instance of the second quantized electronic structure Hamiltonian discretized using Gaussian basis functions. We study this problem in a minimal basis set, which exhibits false minima that can trap the standard variational algorithm despite the problem’s small size. We show that augmenting the training with quantum parallel tempering becomes useful to finding good approximations to the ground states of these problem instances.

arXiv:2210.11405SciPost PhysicsMay 19, 20231 citation

Empowering deep neural quantum states through efficient optimization

Ao Chen, Markus Heyl

Abstract Computing the ground state of interacting quantum matter is a long-standing challenge, especially for complex two-dimensional systems. Recent developments have highlighted the potential of neural quantum states to solve the quantum many-body problem by encoding the many-body wavefunction into artificial neural networks. However, this method has faced the critical limitation that existing optimization algorithms are not suitable for training modern large-scale deep network architectures. Here, we introduce a minimum-step stochastic-reconfiguration optimization algorithm, which allows us to train deep neural quantum states with up to 10 6 parameters. We demonstrate our method for paradigmatic frustrated spin-1/2 models on square and triangular lattices, for which our trained deep networks approach machine precision and yield improved variational energies compared to existing results. Equipped with our optimization algorithm, we find numerical evidence for gapless quantum-spin-liquid phases in the considered models, an open question to date. We present a method that captures the emergent complexity in quantum many-body problems through the expressive power of large-scale artificial neural networks.

arXiv:2302.01941Nature PhysicsJul 1, 202488 citations

Adaptive neural quantum states: a recurrent neural network perspective

Jake McNaughton, Mohamed Hibat-Allah

Abstract Neural-network quantum states (NQS) are powerful neural-network ansätzes that have emerged as promising tools for studying quantum many-body physics through the lens of the variational principle. These architectures are known to be systematically improvable by increasing the number of parameters. Here we demonstrate an Adaptive scheme to optimize NQSs, through the example of recurrent neural networks (RNNs), using a fraction of the computation cost while reducing training fluctuations and improving the quality of variational calculations targeting ground states of prototypical models in one- and two-spatial dimensions. This Adaptive framework reduces the computational cost through training small RNNs and reusing them to initialize larger RNNs. This work opens up the possibility for optimizing graphical processing unit resources deployed in large-scale NQS simulations.

arXiv:2507.18700Machine Learning Science and TechnologySep 15, 20260 citations

Variational neural and tensor network approximations of thermal states

Sirui Lu, Giacomo Giudice, J. I. Cirac

We introduce a variational Monte Carlo algorithm for approximating finite-temperature quantum many-body systems, based on the minimization of a modified free energy. This approach directly approximates the state at a fixed temperature, allowing for systematic improvement of the expressiveness without accumulating errors from iterative imaginary-time evolution. We employ a variety of trial states—both tensor networks as well as neural networks—as variational for our numerical optimization. We benchmark and compare different constructions in the above classes, both for one- and two-dimensional problems, with systems made of up to N = 100 spins. Our results demonstrate that while restricted Boltzmann machines show limitations, string bond tensor network states exhibit systematic improvements with increasing bond dimensions and the number of strings.

arXiv:2401.14243Physical review. B./Physical review. BFeb 3, 20256 citations

Make life simple: unleash the full power of the parallel tempering algorithm

Elmar Bittner, Andreas Nußbaumer, Wolfhard Janke E

We introduce a new update scheme to systematically improve the efficiency of parallel tempering Monte Carlo simulations by taking into account the temperature dependence of autocorrelation times.In contrast to previous attempts the temperatures are not dynamically adjusted but chosen in such a way that the acceptance rate for proposed exchanges of all adjacent replica is about 50%.We show that by adapting the number of sweeps between the parallel tempering moves to the canonical autocorrelation time, the average round-trip time of a replica between the lowest and the highest temperatures is significantly decreased and, therefore, the efficiency of the parallel tempering algorithm is considerably improved.We illustrate the new algorithm with results for the two-dimensional Ising model and propose a toy model to find the optimal parameter set for the parallel tempering routine within no time.

International Symposium on Lattice Field Theory (LATTICE)Jun 23, 20100 citations

Artificial neural network encoding of molecular wavefunctions for quantum computing

Masaya Hagai, Mahito Sugiyama, Koji Tsuda +1 more

Artificial neural networks (ANNs) for material modeling have received significant interest. We recently reported an adaptation of ANNs based on Boltzmann machine (BM) architectures to an ansatz of the multiconfigurational many-electron wavefunction, designated neural-network quantum state (NQS), for quantum chemistry calculations. Here, this study presents its extended formalism to a quantum algorithm that enables the preparation of the NQS through quantum gates. The descriptors of the ANN model, which are chosen as occupancies of electronic configurations, are quantum-mechanically represented by qubits. Our algorithm may thus bring potential advantages over classical sampling-based computation employed in the previous studies. The NQS can be accurately formed through the quantum-native procedures, but the training of the model in terms of energy minimization is performed on a classical computer; thus, our approach is a class of variational quantum eigensolver. The BM models are related to the Gibbs distribution, and our preparation procedures exploit techniques of quantum phase estimation but with no Hamiltonian evolution. The proposed algorithm is assessed by implementing it on a quantum computer simulator. Illustrative molecular calculations at the complete-active-space configuration interaction level of theory are displayed, confirming consistency with the accuracy of our previous classical approaches.

Digital DiscoveryJan 1, 20235 citations