Pith. sign in

REVIEW 4 major objections 5 minor 22 references

Quantum Active Learning for Structural Determination of Doped Nanoparticles -- a Case Study of 4Al@Si$_{11}$

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper claims that a quantum active learning loop built on quantum Gaussian process regression guided DFTB structure searches to the known global minimum of 4Al@Si11 in all 10 independent runs.

desk verdict Near-exhaustive search budget makes the central feasibility claim nearly forced; the paper is an honest case study but needs a random baseline to support its conclusion. read the letter →

arxiv 2412.00504 v1 pith:CEMIE36C submitted 2024-11-30 quant-ph cond-mat.mtrl-sci

Maicon Pierre Lourenço , Mosayeb Naseri , Lizandra Barrios Herrera , Hadi Zadeh-Haghighi , Daya Gaur , Christoph Simon , Dennis R. Salahub This is my paper · ORCID
classification quant-phcond-mat.mtrl-sci
keywords quantumactivelearningGaussianprocessregressiondopednanoparticlesglobalminimumstructuresearchkernelsMBTRdescriptorDFTBstructuraldetermination
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Quantum active learning (QAL) is proposed as a way to find the most stable structure of a doped nanoparticle without exhaustively computing every possibility. The paper's agent is a quantum Gaussian process regression model that, at each cycle, chooses which unobserved dopant arrangements to compute next with density-functional tight-binding (DFTB), then retrains on the new energies. The test bed is 4Al@Si11, an 11-silicon cluster doped with four aluminum atoms, whose 330 arrangements (homotops) all have known DFTB energies, so every search can be scored against a known global minimum. Starting from 20 random structures with energies far above the global minimum, the QAL loop found the putative global minimum in all 10 independent runs. The claim is feasibility: a quantum kernel-based regression can steer structural search in a data-scarce setting, even where it does not beat classical baselines.

What carries the argument

The load-bearing object is the quantum active learning loop: a cycle in which a quantum Gaussian process regressor acts as a decision-making agent, selecting from the unexplored space of homotops the next candidates to be evaluated by DFTB local optimization, after which the observed energies are added to the training set and the regressor is refit. The regressor's similarity measures are quantum kernels — a fidelity quantum kernel $k_{FQK}(\boldsymbol{x}_i,\boldsymbol{x}_j) = \langle \varphi(\boldsymbol{x}_i)|\varphi(\boldsymbol{x}_j)\rangle$ and a projected quantum kernel $k_{PQK}(\boldsymbol{x}_i,\boldsymbol{x}_j) = \exp(-\gamma \sum_{k,P} \{ \mathrm{tr}[P\rho_k(\boldsymbol{x}_i)] - \mathrm{tr}[P\rho_k(\boldsymbol{x}_j)]\}^2)$ built from single-qubit reduced density matrices — produced by two data-encoding circuits, YZ_CX and HighDim. Each homotop is described by the many-body tensor representation (MBTR), reduced by principal component analysis to 4 or 8 components, which fixes the number of qubits. The loop closes by updating the database and repeating until a preset number of cycles is reached; replacing QGPR by a classical Gaussian process regressor defines the classical active learning baseline.

What would settle it

Compute a kernel-target alignment, such as centered kernel alignment, between the FQK and PQK kernel matrices on all 330 4Al@Si11 homotops and the vector of DFTB energies. Near-zero alignment would mean the kernel's notion of similarity carries essentially no energy information, implying the QAL loop's success came from its search dynamics rather than from the quantum kernel. A complementary check: rerun the same QAL loop with the energy labels randomly permuted; if the loop still appears to reach the global minimum under shuffled labels, it is not using the label information at all.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that a quantum active learning loop built on quantum Gaussian process regression (QGPR) is feasible for automatic structural determination of point-defect materials. Concretely, QAL using the projected quantum kernel (PQK) with the YZ_CX feature map performed best among the quantum variants and found the putative global minimum of 4Al@Si11 in all 10 independent runs, even though each run started from 20 random homotops with energies at or above -12.2400 Hartree, deliberately far from the global minimum. With 4-qubit circuits the quantum and classical searches were competitive; with 8 qubits the classical Gaussian process with a dot-product kernel reached the global minimum fastest (in about 40 new calculations), yet the QAL variants still converged there. The authors note throughout that quantum kernel hyperparameters were kept fixed while classical kernel hyperparameters were automatically re-optimized as data accumulated, and they ascribe part of the classical advantage to that asymmetry. The whole demonstration is carried out in a noise-free quantum computing framework, so the claim is about the method's feasibility, not about hardware performance or quantum advantage.

Load-bearing premise

The entire search presupposes that, after MBTR encoding and PCA compression, structures that the quantum kernel judges similar actually have similar DFTB energies; if the compressed descriptors scramble the energy ordering, the quantum agent's selections are barely more informative than random picks, and the loop's success would not be attributable to the learning.

Editorial extensions

If this is right

  • The paper's claim implies that an autonomous quantum-agent loop can replace exhaustive enumeration for small doped clusters: given a descriptor and an energy method, the loop reaches the known global minimum without visiting all 330 homotops.
  • QAL with the projected quantum kernel and YZ_CX feature map is presented as the best quantum variant, improving when the circuit grows from 4 to 8 qubits; if the claim holds, going to more qubits is a plausible route to better search performance.
  • The same QAL procedure is claimed to transfer to other doped nanoparticles and solids with point defects, since the loop is independent of the energy method (DFT or DFTB) and of the specific cluster.
  • Because all QAL variants found the global minimum even from deliberately poor starting populations, the method is presented as reliable across random initial data selection.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper establishes feasibility, not advantage: its own data show the classical Gaussian process with a dot-product kernel and 8 principal components reached the global minimum in about 40 new calculations, faster than any quantum variant, so a reader should take the result as 'quantum active learning works' rather than 'quantum is better.'
  • The comparison is asymmetric in a way the authors flag: quantum kernel hyperparameters were fixed during the loop while classical kernel hyperparameters were re-optimized as data grew; re-running QAL with the quantum kernel's parameters tuned inside the loop is a direct test of whether the classical edge is inherent or an artifact of the fixed configuration.
  • A quantitative test that would separate descriptor quality from kernel expressivity is measuring the alignment between each quantum kernel and the DFTB energy labels on the full 330-homotop set; that analysis is absent from the paper and would explain why PQK outperformed FQK.
  • The sensible next application is a homotop space too large to enumerate fully (bigger clusters, more dopants, or vacancy sites), where the metric is total DFTB effort to reach an energy threshold; that is where a scarce-data search loop could show real value.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes a quantum active learning (QAL) workflow for automatic structural determination of doped nanoparticles, combining quantum Gaussian process regression (QGPR) with fidelity and projected quantum kernels and two feature maps (YZ_CX and HighDim), implemented in the QMLMaterial software. The method is demonstrated on the 4Al@Si11 system, which has 330 DFTB-computed homotops, using 10 independent runs that each start from 20 random high-energy homotops and then perform 60 cycles of 5 new DFTB local optimizations (300 new calculations per run). The central claim, stated in Section 4, is that QAL found the putative global minimum in all 10 runs, demonstrating feasibility for structural determination of point-defect materials. The paper also compares QAL with classical active learning using Gaussian processes with two classical kernels.

Significance. If the claim were supported by a controlled comparison, the paper would be a useful demonstration of a quantum machine learning agent driving a materials discovery loop. The authors ship an implemented workflow (QMLMaterial) and use a concrete, reproducible-looking benchmark system with a known database of 330 homotops. The comparison of FQK and PQK quantum kernels, and of 4-qubit versus 8-qubit encodings, is a reasonable start toward understanding which quantum kernel designs help in active learning. However, the current experimental design does not provide evidence that the quantum acquisition function is actually guiding the search, because the budget is nearly exhaustive and no random-selection baseline is reported. As it stands, the paper is a proof-of-concept of the software plumbing rather than a demonstration of QAL as an effective search strategy.

major comments (4)
  1. [Section 3.1, Figs. 4 and 5] The search budget makes the central claim nearly tautological. With Ncycles = 60 and Nselected = 5, each run performs 300 new DFTB local optimizations on a database of 330 homotops after removing the 20 initial structures. A uniform random rule that never revisits a structure would encounter the global minimum with probability about 300/310 ≈ 0.97 in a single run, and about 0.72 across all 10 runs. Therefore the observation that QAL found the GM in all 10 runs does not distinguish an informative acquisition function from blind enumeration. The paper needs a random-selection baseline and, preferably, a first-hit curve showing the number of new calculations required to reach the GM in each run; without these, the headline result is compatible with the acquisition function being completely ineffective.
  2. [Table 1 and Section 2.5] The quantum kernel hyperparameters (including sigma = 0.0001 for PQK) and the feature-map/PCA choices were selected by grid search on the same fixed 4Al@Si11 energy database that is later used to evaluate QAL. This makes the reported MAE values and the eventual search performance partly fitted outcomes rather than independent predictions. The best-performance claims for QGPR-YZ_CX-PQK should be supported by a model-selection procedure that separates the data used for hyperparameter tuning from the data used for evaluation, or at least by a sensitivity analysis over the kernel hyperparameters.
  3. [Sections 2.1 and 3.1] The paper does not validate whether the MBTR descriptor reduced by PCA to 4 or 8 components preserves the information needed for the quantum kernel to correlate with DFTB energies. No kernel-target alignment, distance-energy correlation, or other descriptor-quality check is reported before the descriptors are used to drive the active learning loop. Given that the random-budget argument shows the current experiment cannot detect a blind kernel, the paper should add a diagnostic that demonstrates the quantum kernel similarities carry information about the energy ordering, otherwise the QAL curves in Figs. 4 and 5 cannot be interpreted as evidence of learning.
  4. [Figures 4 and 5] The comparison between methods is made visually from average-energy curves without error bars or statistical significance tests. With 10 independent runs, the differences between some curves (for example, the QAL-PQK and QAL-FQK curves in Fig. 4) may be within run-to-run variability. The authors should report standard deviations or confidence intervals, and ideally a paired test across runs when comparing methods on the same initial conditions.
minor comments (5)
  1. [Section 2.2, Eq. (3)] The RBF kernel is introduced with the sentence 'The RBF kernel is given by Eq. 1' but the displayed formula is numbered Eq. (3); the cross-reference should be corrected.
  2. [Section 2.3, Eq. (5)] The projected quantum kernel formula contains an unfinished expression 'ρ_k(x_i) =' immediately before the text; the definition of the 1-RDM and the trace operator should be written out completely.
  3. [Section 3.1] The circuit name is written inconsistently as 'YC_ZX' in one place and 'YZ_CX' in others; also Fig. 3 is referenced in the text as 'Fig. X' and 'Fig. Y' in Section 2.5, and the figure captions should be checked.
  4. [Section 2.5 and Table 1] The text alternates between 'Tab. 1' and 'Tab. 2' for what appears to be the same hyperparameter table; the numbering and the associated explanations should be unified.
  5. [Throughout] There are several typographical issues, including 'Hatree' for Hartree, 'strutural' for structural, '4AL@Si11' for 4Al@Si11, and duplicated or malformed references in the reference list; a careful proofreading pass is needed.

Circularity Check

2 steps flagged · score 6.0 of 10

QAL feasibility claim is largely forced by a near-exhaustive 300-of-330 search budget and by hyperparameters tuned on the same benchmark database.

  1. fitted input called prediction [Section 2.5 / Table 1 caption]
    "Tab. 1 presents the mean absolute error (MAE) for the training and testing set of 4Al@Si11 obtained by different QAL and AL iterations. Also, the set up used to obtain the optimum hyperparameters is presented. They were obtained by grid search for a fixed 4Al@Si11 energy database."

    The quantum kernel hyperparameters (e.g., sigma = 0.0001, PQK with HighDim/YZ_CX feature maps) were selected by minimizing MAE on the same fixed 330-homotop 4Al@Si11 energy database that is later used to claim that QAL 'found the putative GM'. Because the model was tuned on the target benchmark, the reported QAL success on 4Al@Si11 is not an independent prediction; the kernel parameters were fitted to the very energy landscape it is then said to discover.

  2. other [Section 3.1 / Fig. 4 caption; Section 4 Conclusions]
    "Average total energy (in Hartree) obtained by DFTB (in Hartree) for 4Al@Si11 obtained by 10 QAL and AL independent runs as a function of new calculations (with Ncycles = 60 and Nselected = 5). ... In all cases, the QAL found the putative GM of 4Al@Si11, showing that the proposed QAL is feasible for automatic structural determination of point-defect materials."

    With Ncycles = 60 and Nselected = 5, each run performs 300 new DFTB local optimizations on a fixed database of only 330 homotops, after starting from 20 initial random structures. Therefore about 310 unexplored homotops remain and the QAL visits 300 of them, i.e., roughly 97% of the whole search space. Even a purely random acquisition without replacement would find the GM with probability 300/310 = 96.8% per run, and about 72% across all 10 runs. The conclusion that QAL 'found the GM in all cases' is thus forced by the nearly exhaustive budget rather than by the quantum acquisition function; the reported result is compatible with the active-learning model being completely ineffective as a search guide.

full rationale

The paper's active-learning loop is implemented in code and the comparison between QGPR and classical GPR is real benchmarking, so the derivation is not identical to its inputs. However, two load-bearing aspects of the claimed demonstration are circular in the specific sense defined by the protocol. First, Table 1 states that the optimum QGPR/GPR hyperparameters 'were obtained by grid search for a fixed 4Al@Si11 energy database,' and these same hyperparameters were then used in the QAL runs on that same 330-homotop benchmark; the reported success on 4Al@Si11 is therefore an in-sample, hyperparameter-fitted result rather than an independent prediction. Second, and more decisively, the protocol uses Ncycles = 60 and Nselected = 5, i.e., 300 new local optimizations per run on a database of 330 homotops, starting from 20 initial structures. That leaves only about 310 unexplored homotops, of which the QAL visits about 300; a random acquisition without replacement would find the GM with probability ~96.8% per run and ~72% over 10 runs. The conclusion 'In all cases, the QAL found the putative GM ... showing that the proposed QAL is feasible' is thus statistically forced by the near-exhaustive budget and cannot support the feasibility claim without a random baseline or a time-to-first-GM comparison. The self-citations to Refs. 2, 20, and 26 are a normal use of prior work and are not load-bearing circularity by themselves.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

No new physical entities are introduced; the paper builds a software workflow from existing QML algorithms. The main burdens are the descriptor choice, the DFTB oracle, and the search protocol, plus hyperparameters tuned on the evaluation set.

free parameters (4)
  • sigma_PQK = 0.0001
    Regularization hyperparameter for the projected quantum kernel; selected by grid search on the same 4Al@Si11 energy database used for evaluation (Section 2.5, Table 1).
  • PCA dimension = 4 or 8
    Number of principal components chosen to match the qubit count; no independent validation that this preserves energy-relevant structure (Sections 3.1 and 3.2).
  • Search budget = Ncycles=60, Nselected=5, initial N=20
    Experiment design choice; covers up to 320 of 330 homotops, making global minimum discovery near-guaranteed (Figure 4 and 5 captions).
  • Quantum circuit choices = YZ_CX, HighDim with reps=4
    Feature maps and circuit repetitions chosen based on mean absolute error on the same system; no external benchmark (Sections 2.3 and 2.5).
assumptions (5)
  • domain assumption DFTB energies for all 330 homotops, taken from Ref 2, are an accurate ground truth for the global minimum search.
    The benchmark treats DFTB as the oracle; inaccuracies in the DFTB parameterization would make the 'global minimum' not the true physical minimum (Section 3).
  • domain assumption MBTR descriptor plus PCA retains enough structural information for kernel-based energy ranking.
    All kernels operate on the reduced descriptor; no direct validation of the embedding is given (Sections 2.1 and 3.1).
  • domain assumption The noise-free quantum circuit simulations in sQUlearn faithfully compute the fidelity and projected quantum kernels.
    The method is formulated in a noise-free framework, and hardware noise is not modeled (Abstract and Section 2.3).
  • domain assumption Exploitation-only acquisition, selecting the lowest predicted energy, is an effective strategy for the QAL loop.
    The paper states exploitation was used, but does not compare with exploration strategies or an explicit acquisition function (Section 3.1 and Figure 4 caption).
  • domain assumption The 330 enumerated homotops are complete for 4Al@Si11.
    The method searches only this precomputed set; any missing structure would be invisible to the benchmark (Section 1 and Figure 1B).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Quantum Active Learning for Structural Determination of Doped Nanoparticles -- a Case Study of 4Al@Si$_{11}$." pith.science (2026). https://pith.science/paper/CEMIE36C

@misc{pith2026241200504,
  author       = {Pith},
  title        = {Pith review of: Quantum Active Learning for Structural Determination of Doped Nanoparticles -- a Case Study of 4Al@Si$_11$},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CEMIE36C}},
  note         = {Machine review of arXiv:2412.00504}
}
abstract

Active learning (AL) has been widely applied in chemistry and materials science. In this work we propose a quantum active learning (QAL) method for automatic structural determination of doped nanoparticles, where quantum machine learning (QML) models for regression are used iteratively to indicate new structures to be calculated by DFT or DFTB and this new data acquisition is used to retrain the QML models. The QAL method is implemented in the Quantum Machine Learning Software/Agent for Material Design and Discovery (QMLMaterial), whose aim is using an artificial agent (defined by QML regression algorithms) that chooses the next doped configuration to be calculated that has a higher probability of finding the optimum structure. The QAL uses a quantum Gaussian process with a fidelity quantum kernel as well as the projected quantum kernel and different quantum circuits. For comparison, classical AL was used with a classical Gaussian process with different classical kernels. The presented QAL method was applied in the structural determination of doped Si$_{11}$ with 4 Al (4Al@Si$_{11}$) and the results indicate the QAL method is able to find the optimum 4Al@Si$_{11}$ structure. The aim of this work is to present the QAL method -- formulated in a noise-free quantum computing framework -- for automatic structural determination of doped nanoparticles and materials defects.

Figures

Figures reproduced from arXiv: 2412.00504 by the authors.

Figure 1
Figure 1. (A) Putative global minimum structure (GM) of 4Al@Si11 found by DFTB and active learning. (B) DFTB energy distribution (in Hartree) of the 330 4Al@Si11 homotops or isomers. This paper is organized as follows: the next section introduces the methods employed in this work, providing a brief overview of active learning, the concepts of classical and quantum kernels as well as classical and quantum ML set up. Section 3 … view at source ↗
Figure 2
Figure 2. Artificial intelligence (AI) workflow based on quantum and classical active learning (QAL and AL) for optimum experimental design. This is implemented in QMLMaterial software using sQUlearn quantum framework. 2.2 Classical kernels Kernel based methods for regression – as in classical GP – are within a model widely employed in supervised learning. It requires a prior’s covariance that has to be specified by passing a… view at source ↗
Figure 1
Figure 1. More details about the [PITH_FULL_IMAGE:figures/full_fig_p014_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

22 extracted references · 15 canonical work pages

  1. [1]

    S.; Larsen, U

    (1) Jørgensen, M. S.; Larsen, U. F.; Jacobsen, K. W.; Hammer, B. Exploration versus exploitation in global atomistic structure optimization. J Phys Chem A 2018,

    work page 2018

  2. [5]

    (24) Hohenberg, P.; Kohn, W

    DOI: 10.1007/s42484-023-00138-9. (24) Hohenberg, P.; Kohn, W. Inhomogeneous Electron Gas. Physical Review 1964, 136 (3B), B864- B871. DOI: 10.1103/PhysRev.136.B864. (25) Koskinen, P.; Mäkinen, V . Density -functional tight -binding for beginners. Computational Materials Science 2009, 47 (1), 237-253. DOI: https://doi.org/10.1016/j.commatsci.2009.07.013. (...

    work page doi:10.1007/s42484-023-00138-9 1964

  3. [15]

    (20) Lourenço, M.; Za deh-Haghighi, H.; Hostaš, J.; Naseri, M.; Gaur, D.; Simon, C.; Salahub, D

    DOI: 10.48550/arXiv.2306.16028. (20) Lourenço, M.; Za deh-Haghighi, H.; Hostaš, J.; Naseri, M.; Gaur, D.; Simon, C.; Salahub, D. Exploring Quantum Active Learning for Materials Design and Discovery

    arXiv doi:10.48550/arxiv.2306.16028

  4. [16]

    (21) Ding, Y.; Ban, Y.; Sanz, M.; Martín-Guerrero, J

    DOI: 10.26434/chemrxiv-2024-kt165. (21) Ding, Y.; Ban, Y.; Sanz, M.; Martín-Guerrero, J. D.; Chen, X. Quantum Active Learning. arXiv preprint arXiv:2405.18230

    work page arXiv 2024

  5. [17]

    (23) Rapp, F.; Roth, M

    DOI: 10.48550/arXiv.2409.04406. (23) Rapp, F.; Roth, M. Quantum Gaussian process regression for Bayes ian optimization. Quantum Machine Intelligence 2024, 6 (1),

    arXiv doi:10.48550/arxiv.2409.04406 2024

  6. [35]

    (6) Lourenco, M

    DOI: 10.1038/s41524 -019- 0175-2. (6) Lourenco, M. P.; Tchagang, A.; S hankar, K.; Thangadurai, V.; Salahub, D. R. Active Learning for Optimum Experimental Design – Insight into Perovskite Oxides. Canadian Journal of Chemistry

    doi:10.1038/s41524

  7. [37]

    (31) Pedregosa, F.; Ga; #235; Varoquaux, l.; Gramfort, A.; Michel, V.; Thiri on, B.; Grisel, O.; Blondel, M.; Prettenhofer, P.; et al

    DOI: 10.1038/s41524-018-0096-5. (31) Pedregosa, F.; Ga; #235; Varoquaux, l.; Gramfort, A.; Michel, V.; Thiri on, B.; Grisel, O.; Blondel, M.; Prettenhofer, P.; et al. Scikit -learn: Machine Learning in Python. J. Mach. Learn. Res. 2011, 12, 2825-2830. (32) Rossum, G. V.; Drake, F. L. Python 3 Reference Manual; CreateSpace,

    work page doi:10.1038/s41524-018-0096-5

  8. [62]

    (3) Lourenço, M

    DOI: 10.1007/s00214-021-02766-5. (3) Lourenço, M. P.; dos Santos Anastácio, A.; Ros a, A. L.; Frauenheim, T.; da Silva, M. C. An adaptive design approach for defects distribution modeling in materials from first -principle calculations. Journal of Molecular Modeling 2020, 26 (7),

    work page doi:10.1007/s00214-021-02766-5

Show all 22 references
  1. [116]

    (2) Lourenço, M

    DOI: 10.1007/s00214-021-02820-2. (2) Lourenço, M. P.; Galvão, B. R. L.; Barrios Herrera, L.; Hostaš, J.; Tchagang, A.; Silva, M. X.; Salahub, D. R. A new active learning approach for global optimization of atomic clusters. Theoretical Chemistry Accounts 2021, 140 (6),

    doi

  2. [122]

    Bisbo, M

    DOI: 10.1021/acs.jpca.8b00160. Bisbo, M. K.; Hammer, B. Efficient Global Structure Optimiz ation with a Machine -Learned Surrogate Model. Physical Review Letters 2020, 124 (8), 086102. DOI: 10.1103/PhysRevLett.124.086102. Lourenço, M. P.; Herrera, L. B.; Hostaš, J.; Calaminici...

    2020 doi

  3. [161]

    (18) Haug, T.; Self, C

    DOI: 10.1038/s41534-021-00498-9. (18) Haug, T.; Self, C. N.; Kim, M. S. Quantum machine learning of large datasets using randomized measurements. Machine Learning: Science and Technology 2023, 4 (1), 015005. DOI: 10.1088/2632- 2153/acb0b4. (19) Gyurik, C.; Dunjko, V. Exponenti...

    2023 doi

  4. [164]

    (7) Jolliffe, I

    DOI: 10.1016/j.commatsci.2019.03.057. (7) Jolliffe, I. T.; Cadima, J. Principal component analysis: a review and recent developments. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 2016, 374 (2065), 20150202. DOI: doi:10.1098...

    2019

  5. [178]

    (5) Todorović, M.; Gutmann, M

    DOI: 10.1007/s00894-022-05173-0. (5) Todorović, M.; Gutmann, M. U.; Corander, J.; Rinke, P. Bayesian inference of atomistic structure in functional materials. npj Computational Materials 2019, 5 (1),

    2019 doi

  6. [187]

    Lourenço, M

    DOI: 10.1007/s00894-020-04438-w. Lourenço, M. P.; Herrera, L. B.; Hostaš, J.; Calaminici, P.; Köster, A. M.; Tchagang, A.; Salahub, D. R. Automatic structural elucidation of vacancies in materials by active learning. Physical Chemistry Chemical Physics 2022, 10.1039/D2CP02585J...

    doi

  7. [250]

    (35) Hourahine, B.; Aradi, B.; Blum, V.; B onafé, F.; Buccheri, A.; Camacho, C.; Cevallos, C.; Deshaye, M

    DOI: 10.1007/s00214-016-2001-y. (35) Hourahine, B.; Aradi, B.; Blum, V.; B onafé, F.; Buccheri, A.; Camacho, C.; Cevallos, C.; Deshaye, M. Y.; Dumitrică, T.; Dominguez, A.; et al. DFTB+, a software package for efficient approximate density functional theory based atomistic sim...

    2001 doi

  8. [303]

    (34) Lourenço, M

    DOI: 10.1007/s00894-020-04484-4. (34) Lourenço, M. P.; da Silv a, M. C.; Oliveira, A. F.; Quintão, M. C.; Duarte, H. A. FASP: a framework for automation of Slater –Koster file parameterization. Theoretical Chemistry Accounts 2016, 135 (11),

    doi

  9. [720]

    (16) Zhu, D.; Linke, N

    DOI: 10.22331/q-2022-05-24-720. (16) Zhu, D.; Linke, N. M.; Benedetti, M.; Landsman, K. A.; Nguyen, N. H.; Alderete, C. H.; Perdomo-Ortiz, A.; Korda, N.; Garfoot, A.; Brecque, C.; et al. Training of quantum circuits on a hybrid quantum computer. Science Advances 2019, 5 (10), ...

    2022 doi

  10. [2009]

    (33) Galvão, B. R. L.; Viegas, L. P .; Salahub, D. R.; Lourenço, M. P. Reliability of semiempirical and DFTB methods for the global optimization of the structures of nanoclusters. Journal of Molecular Modeling 2020, 26 (11),

    2020

  11. [2021]

    Quantum agents in the Gym : a variational quantum algorithm for deep Q-learning

    (15) Skolik, A.; Jerbi, S.; Dunjko, V. Quantum agents in the Gym : a variational quantum algorithm for deep Q-learning. Quantum 2022, 6,

    2022

  12. [2023]

    Balachandran, P

    DOI: 10.1139/cjc-2022-0198 (acccessed 2023/05/10). Balachandran, P. V. Machine learning guided design of functional materials with targeted properties. Comput Mater Sci 2019,

    2022 doi

  13. [2024]

    (10) Biamonte, J.; Wittek, P.; Pancotti, N.; Rebentrost, P.; Wiebe, N.; Lloyd, S

    DOI: 10.48550/arXiv.2311.08990. (10) Biamonte, J.; Wittek, P.; Pancotti, N.; Rebentrost, P.; Wiebe, N.; Lloyd, S. Quantum machine learning. Nature 2017, 549 (7671), 195-202. DOI: 10.1038/nature23474. Dunjko, V.; Briegel, H. J. Machine learning & artificial intelligence in ...

    2017 doi arXiv

  14. [2631]

    (9) Kreplin, D.; Willmann, M.; Schnabel, J.; Rapp, F.; Roth, M

    DOI: 10.1038/s41467-021-22539-9. (9) Kreplin, D.; Willmann, M.; Schnabel, J.; Rapp, F.; Roth, M. sQUlearn – A Python Library for Quantum Machine Learning

    doi

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.