Pith. sign in

REVIEW 4 major objections 5 minor 58 references

Efficient training and design of photonic neural network through neuroevolution

T0 review · 4 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Gradient-free evolution trains photonic neural networks competitively

desk verdict A credible but limited demonstration that evolutionary algorithms can train simulated optical neural networks; the accuracy/stability competitiveness claim needs more runs and baselines. read the letter →

arxiv 1908.08012 v1 pith:WBDBJLOL submitted 2019-08-04 cs.NE physics.optics

Tian Zhang , Jia Wang , Yihang Dan , Yuxiang Lanqiu , Jian Dai , Xu Han , Xiaojuan Sun , Kun Xu This is my paper · ORCID
classification cs.NEphysics.optics
keywords photonicneuralnetworksneuroevolutiongeneticalgorithmparticleswarmoptimizationMach-Zehnderinterferometerphaseshifterselectro-opticactivationgradient-freetraining
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

The paper claims that neuroevolution—genetic algorithms and particle swarm optimization—can train photonic neural networks by directly optimizing the phase-shifter weights of optical meshes and the hyperparameters of electro-optic activation functions. In chip-level simulations, these gradient-free methods reach classification accuracies comparable to gradient-based training such as the adjoint variable method on iris, wine, and modulation-format recognition tasks. If the claim holds, optical neural networks would no longer need to map pre-trained weights from digital computers or compute gradients in situ; they could be designed and trained by evolutionary search alone. This matters because gradient computation is awkward in integrated photonic hardware, while evolutionary search only needs a scalar fitness signal.

What carries the argument

The central object is the programmable Mach-Zehnder interferometer mesh, whose phase shifters implement arbitrary unitary matrices via standard decompositions, followed by electro-optic activation functions whose response is set by three physical hyperparameters: tapped power fraction $\alpha$, phase gain $g$, and biasing phase $\theta$. The neuroevolution algorithms treat both the phase-shifter values and these hyperparameters as a single optimization vector, using classification loss as fitness. The genetic algorithm converts variables to binary for crossover and mutation, while particle swarm optimization operates directly on decimal values, which the paper argues avoids precision loss and explains its better accuracy.

What would settle it

Train an ONN on a real integrated photonic chip using the same evolutionary search, or add measured hardware noise (phase error, insertion loss, activation mismatch) to the simulator and re-run the iris and wine tasks; if accuracies drop substantially below the simulated results or fail to converge, the claim of competitiveness with traditional training on physical ONNs is refuted.

Watch

Extended reading notes

Core claim

The central claim is that neuroevolution is a viable training strategy for optical neural networks, competitive with traditional learning algorithms on accuracy and stability. In chip-level simulations, the authors train networks whose linear layers are Mach-Zehnder interferometer meshes and whose nonlinear layers are electro-optic activation functions controlled by three physical parameters. The genetic algorithm and particle swarm optimization evolve both the phase-shifter weights and the activation hyperparameters, and on three benchmark classification tasks the particle swarm optimizer reaches 100% accuracy on iris and wine datasets and 93% on modulation-format recognition. The paper positions this as an efficient, gradient-free alternative to the adjoint variable method and stochastic gradient descent for photonic neural networks.

Load-bearing premise

The entire demonstration runs in the paper's chip-level simulation platform, so the claim that neuroevolution trains photonic neural networks depends on the simulator faithfully representing a physical photonic chip; if real phase errors, amplitude drift, or activation nonlinearity mismatches break the trained weights, the method's practical value is unproven.

Editorial extensions

If this is right

  • Optical neural networks can in principle be trained in situ with only a scalar loss signal, removing the need for backpropagation or gradient measurement through photonic hardware.
  • Network architecture and activation hyperparameters can be co-optimized with weights, so the trained object is the whole hardware configuration rather than a weight matrix mapped onto a chip.
  • The same evolutionary training could extend to deep reinforcement learning with photonic networks, since evolutionary strategies are already competitive with policy-gradient methods in that setting.
  • Population-based search is parallelizable across many candidate chips or simulations, potentially speeding up training when gradient evaluation is costly or unavailable.

Reading between the lines

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

  • A natural extension is noise-aware neuroevolution: injecting measured fabrication and thermal phase errors during fitness evaluation could evolve configurations that are robust on real hardware, something the paper does not test.
  • Because particle swarm optimization outperforms the genetic algorithm here largely due to avoiding binary encoding, a continuous evolutionary strategy might push accuracy further on larger meshes.
  • The method's scaling to deep networks with hundreds of phase shifters is untested; population size and iteration limits suggest the approach may need hybridization with gradient-based local refinement for large-scale optical neural networks.
  • The same fitness-based training could be applied to other programmable photonic circuits, such as mesh-based linear transformers for optical signal processing, where no gradient channel exists.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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 manuscript proposes using genetic algorithms (GA) and particle swarm optimization (PSO) to train simulated photonic neural networks implemented as Mach-Zehnder interferometer meshes with electro-optic nonlinear activation functions. The algorithms optimize both the phase-shifter weights and three activation hyper-parameters (α, g, θ) from Ref. [31]. Training is evaluated on two synthetic binary-classification datasets from the neuroptica simulator, with the adjoint variable method (AVM) as the comparison baseline, and on iris, wine, and modulation-format recognition tasks. The authors report convergence of training losses and classification accuracies, leading to the abstract claim that neuroevolution is competitive with traditional learning algorithms on both accuracy and stability.

Significance. If validated, the paper would provide a gradient-free training route for optical neural networks that simultaneously optimizes device-level hyperparameters—useful in settings where in situ gradient measurement is difficult. The study has clear strengths: it builds on the openly available neuroptica simulator, optimizes physically meaningful activation parameters, covers multiple datasets, and includes ablations on population size and selection operators. However, the evidence as presented is not yet sufficient to support the headline accuracy/stability claim; the missing statistical replication and dataset-split details are fixable within the paper's scope. The significance for hardware deployment remains prospective because no experimental validation or hardware-error analysis is provided.

major comments (4)
  1. [Section 3, Figs. 3(b)-4(d)] The central claim that neuroevolution is 'competitive with other traditional learning algorithms on both accuracy and stability' is not supported by the reported experiments because every training curve appears to be a single run. GA and PSO are stochastic (Section 2: roulette-wheel/tournament selection, uniform crossover, 5% bit-flip mutation; random r1 and r2 in Eq. (2)), so run-to-run variation is expected. No seeds, repetitions, error bars, or variance estimates are reported, and the term 'stability' is never defined. Please add multi-seed repetitions and report mean and standard deviation or confidence intervals, and define the stability metric explicitly.
  2. [Section 3, iris/wine/modulation-format experiments] The train/test split is specified only for the synthetic neuroptica datasets (80% training, 20% test). For the iris, wine, and modulation-format datasets, the manuscript reports 'test dataset' accuracies (Figs. 3(c-d) and 4(c-d)) without stating how instances were partitioned, whether features were normalized, or whether the reported numbers are training or test accuracy. Without this information, the accuracy values are not reproducible and the comparison is uninterpretable.
  3. [Section 3, Figs. 3(b) and 4(b)] The comparison with 'other traditional learning algorithms' is limited to AVM on two synthetic datasets. On the triangle dataset the GA ends at MSE 0.09 versus 0.005 and accuracy 0.95 versus 1.00 for AVM, so 'competitive' is effectively demonstrated only by PSO on the ring dataset; PSO is not compared with AVM or SGD/Adam on any of the three real datasets. The abstract and conclusions should be qualified accordingly, or the missing comparisons should be added.
  4. [Section 2 and Section 3] All results are obtained in the neuroptica simulation platform using the electro-optic activation model of Ref. [31]. The manuscript does not discuss how fabrication errors, phase drift, amplitude nonuniformity, or nonlinearity mismatch would affect the optimized weights and hyper-parameters. The conclusion that this is an 'efficient training method for the ONNs' should be scoped to simulation-level demonstration, or supported by a hardware-aware error analysis.
minor comments (5)
  1. [Eq. (2)] The inertia weight W in the PSO velocity update is never assigned a value; please specify it together with the velocity clamping range (-2 to 2).
  2. [Section 2 vs. Section 3] The stopping criterion is stated as 1000 generations in Section 2, but Section 3 reports GA results 'after 2000 iterations'; please reconcile these numbers.
  3. [Introduction, paragraph 2] The sentence claiming that neuroevolution outperforms DRL algorithms on Atari references Ref. [42], Hessel et al. (Rainbow), which is not a neuroevolution method; the supporting reference appears to be incorrect.
  4. [Throughout] There are numerous typos and inconsistent abbreviations: 'cross' should be 'crossover', 'train method' should be 'training method', and 'DTR' in the Conclusions should be 'DRT'.
  5. [Fig. 3(b)] The caption mentions classification boundary contours for the GA, but the text does not explain how these contours are obtained; please clarify.

Circularity Check

0 steps flagged · score 0.0 of 10

No derivational circularity: the claims are empirical benchmarks in an external simulator, not predictions forced by fitted inputs.

full rationale

The paper's central claim is that GA/PSO neuroevolution can train simulated ONNs competitively with AVM/SGD. This is tested by running the evolution algorithms on classification datasets (iris, wine, modulation-format) and on two toy datasets from the neuroptica platform, with the AVM as a baseline. The optimized quantities (MZI phase shifters and electro-optic activation hyperparameters alpha, g, theta) are not derived from the target accuracies; the accuracies are measured after optimization. No equation defines the reported error rate in terms of the training objective by construction; the fitness is the loss, but minimizing a loss and then reporting that loss on held-out data is a normal empirical protocol, not a circular reduction. The paper relies on Ref. [31] for the electro-optic activation model and on the neuroptica simulator [47]; both are external to the present authors and are independent support. Ref. [36] is the authors' own prior work, but it is cited only for standard GA selection and crossover details and is not load-bearing for the accuracy/stability claim. There is no self-citation uniqueness theorem, no ansatz smuggled in via citation, and no renaming of a known result. The absence of repeated runs and error bars weakens the stability claim empirically, but that is an evidence-strength issue, not circularity. Therefore the circularity burden is not met and the score is 0.

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

The paper contributes an empirical application of standard evolutionary optimizers to a simulator. The central numbers depend on manually chosen population sizes, layer counts, and algorithm constants, and on the fidelity of the neuroptica model and the electro-optic activation function. No new physical entities are introduced.

free parameters (5)
  • Population size N = 500 for both GA and PSO
    Results in Figs. 3(e) and 4(e) vary with population size; N=500 is used for the final reported accuracies.
  • Network layer count L = 3 for iris and modulation, 2 for wine, 5 for synthetic datasets
    Architecture is chosen by hand per dataset; Fig. 4(f) shows accuracy depends on L, so this choice affects the central results.
  • GA crossover and mutation probabilities = crossover 0.8, mutation 0.05, gene exchange 0.5
    Fixed without sensitivity analysis; these control the evolutionary search dynamics.
  • PSO coefficients = c1=c2=1.49445, velocity limited to [-2,2], inertia weight W not specified
    Equations (2) and (3) define the PSO update; the constants are chosen by hand and no value is given for W.
  • Stopping criteria = 1000 generations or 5 generations without loss change
    Used for all experiments; affects convergence and final accuracy.
assumptions (4)
  • domain assumption The neuroptica simulator faithfully represents an optical neural network, including MZI meshes and electro-optic nonlinearities.
    All training and evaluation happens in simulation (Sections 2-3, Refs. 31 and 47). If the model is inaccurate, the results do not transfer to hardware.
  • standard math Any unitary matrix can be implemented by the MZI mesh via Reck or Clements decomposition.
    Relied on in Section 2 when constructing unitary optical meshes; standard result from Refs. 48-49.
  • domain assumption The electro-optic activation function model with parameters alpha, g, and theta from Ref. 31 accurately describes the nonlinearity.
    The paper optimizes these hyperparameters; if the model is wrong, the optimized values have no physical meaning.
  • domain assumption The four statistical features are sufficient for modulation format recognition.
    Used in Section 3 for the modulation dataset; inherited from Ref. 53 without re-validation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Efficient training and design of photonic neural network through neuroevolution." pith.science (2026). https://pith.science/paper/WBDBJLOL

@misc{pith2026190808012,
  author       = {Pith},
  title        = {Pith review of: Efficient training and design of photonic neural network through neuroevolution},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WBDBJLOL}},
  note         = {Machine review of arXiv:1908.08012}
}
read the original abstract

Recently, optical neural networks (ONNs) integrated in photonic chips has received extensive attention because they are expected to implement the same pattern recognition tasks in the electronic platforms with high efficiency and low power consumption. However, the current lack of various learning algorithms to train the ONNs obstructs their further development. In this article, we propose a novel learning strategy based on neuroevolution to design and train the ONNs. Two typical neuroevolution algorithms are used to determine the hyper-parameters of the ONNs and to optimize the weights (phase shifters) in the connections. In order to demonstrate the effectiveness of the training algorithms, the trained ONNs are applied in the classification tasks for iris plants dataset, wine recognition dataset and modulation formats recognition. The calculated results exhibit that the training algorithms based on neuroevolution are competitive with other traditional learning algorithms on both accuracy and stability. Compared with previous works, we introduce an efficient training method for the ONNs and demonstrate their broad application prospects in pattern recognition, reinforcement learning and so on.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

58 extracted references · 54 canonical work pages

  1. [31]

    On-Chip Optical Convolutional Neural Networks

    H. Bagherian, S. Skirlo, Y. Shen, H. Meng, V. Ceperic, and M. Soljacic, "On-Chip Optical Convolutional Neural Networks," arXiv preprint arXiv:1808.03303 (2018)

    work page Pith review arXiv 2018

  2. [36]

    Reprogrammable Electro-Optic Nonlinear Activation Functions for Optical Neural Networks

    I. A. Williamson, T. W. Hughes, M. Minkov, B. Bartlett, S. Pai, and S. Fan, "Reprogrammable Electro-Optic Nonlinear Activation Functions for Optical Neural Networks," arXiv preprint arXiv:1903.04579 (2019)

    work page Pith review arXiv 2019

  3. [1]

    Introduction Artificial neural networks (ANNs), deep learning [1] in particular, has attracted a great deal of research attentions for an impressively large number of applications, such as image processing [2], natural language processing [3], acoustical signal processing [4], time series processing [5], self-driving [6], games [7], robot [8] and so on. I...

    work page

  4. [2]

    1(a), the network architecture of ANNs imitates the structure of biological neural network which includes a great number of neuron s and connections layer by layer [1]

    Training methods based on neuroevolution As shown in Fig. 1(a), the network architecture of ANNs imitates the structure of biological neural network which includes a great number of neuron s and connections layer by layer [1]. It should noticed that although ANNs are brain -inspired, there are significant differences in the network structure, learning met...

    work page

  5. [3]

    The iris plants dataset is a simple dataset which includes 150 instances (4 attributes for each instance)

    Calculated results and discussions As typical datasets in the classification tasks, the iris plants dataset [51] and wine recognition dataset [52] are selected as the test datasets to demonstrate the effectiveness of the training algorithm. The iris plants dataset is a simple dataset which includes 150 instances (4 attributes for each instance). Compared ...

    work page 2000

  6. [4]

    Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups,

    G. Hinton, L. Deng, D. Yu, G. E. Dahl, A. Mohamed, N. Jaitly, A. Senior, V. Vanhoucke, P. Nguyen, and T. N. Sainath, "Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups," IEEE Signal Processing Magazine 29(6), 82-97 (2012)

    work page 2012

  7. [5]

    A review of unsupervised feature learning and deep learning for time-series modeling,

    M. Lä ngkvist, L. Karlsson, and A. Loutfi, "A review of unsupervised feature learning and deep learning for time-series modeling," Pattern Recognition Letters 42, 11-24 (2014)

    work page 2014

  8. [6]

    Deep learning,

    Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," Nature 521(7553), 436 (2015)

    work page 2015

Show all 58 references
  1. [7]

    Deep learning in neural networks: An overview,

    J. Schmidhuber, "Deep learning in neural networks: An overview," Neural networks 61, 85-117 (2015)

    2015

  2. [8]

    Recent trends in deep learning based natural language processing,

    T. Young, D. Hazarika, S. Poria, and E. Cambria, "Recent trends in deep learning based natural language processing," ieee Computational intelligenCe magazine 13(3), 55-75 (2018)

    2018

  3. [9]

    Imagenet classification with deep convolutional neural networks,

    A. Krizhevsky, I. Sutskever, and G. E. Hinton, "Imagenet classification with deep convolutional neural networks," in Advances in neural information processing systems, (NIPS, 2012), 1097-1105

    2012

  4. [10]

    Very deep convolutional networks for large-scale image recognition,

    K. Simonyan and A. Zisserman, "Very deep convolutional networks for large-scale image recognition," arXiv preprint arXiv:1409.1556 (2014)

    2014 arXiv

  5. [11]

    End to end learning for self-driving cars,

    M. Bojarski, D. Del Testa, D. Dworakowski, B. Firner, B. Flepp, P. Goyal, L. D. Jackel, M. Monfort, U. Muller, and J. Zhang, "End to end learning for self-driving cars," arXiv preprint arXiv:1604.07316 (2016)

    2016 arXiv

  6. [12]

    Playing atari with deep reinforcement learning,

    V. Mnih, K. Kavukcuoglu, D. Silver, A. Graves, I. Antonoglou, D. Wierstra, and M. Riedmiller, "Playing atari with deep reinforcement learning," arXiv preprint arXiv:1312.5602 (2013)

    2013 arXiv

  7. [13]

    Deep reinforcement learning for robotic manipulation with asynchronous off-policy updates,

    S. Gu, E. Holly, T. Lillicrap, and S. Levine, "Deep reinforcement learning for robotic manipulation with asynchronous off-policy updates," in 2017 IEEE international conference on robotics and automation (ICRA), (IEEE, 2017), 3389-3396

    2017

  8. [14]

    Theano: Deep learning on gpus with python,

    J. Bergstra, F. Bastien, O. Breuleux, P. Lamblin, R. Pascanu, O. Delalleau, G. Desjardins, D. Warde-Farley, I. Goodfellow, and A. Bergeron, "Theano: Deep learning on gpus with python," in NIPS 2011, BigLearning Workshop, Granada, Spain, (Citeseer, 2011), 1-48

    2011

  9. [15]

    Optimizing fpga-based accelerator design for deep convolutional neural networks,

    C. Zhang, P. Li, G. Sun, Y. Guan, B. Xiao, and J. Cong, "Optimizing fpga-based accelerator design for deep convolutional neural networks," in Proceedings of the 2015 ACM/SIGDA International Symposium on Field- Programmable Gate Arrays, (ACM, 2015), 161-170

    2015

  10. [16]

    Going deeper with convolutions,

    C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich, "Going deeper with convolutions," in Proceedings of the IEEE conference on computer vision and pattern recognition, (CVPR, 2015), 1-9

    2015

  11. [17]

    Deep residual learning for image recognition,

    K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition," in Proceedings of the IEEE conference on computer vision and pattern recognition, (IEEE, 2016), 770-778

    2016

  12. [18]

    Long short-term memory,

    S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural computation 9(8), 1735-1780 (1997)

    1997

  13. [19]

    The spinnaker project,

    S. B. Furber, F. Galluppi, S. Temple, and L. A. Plana, "The spinnaker project," Proc. IEEE 102(5), 652-665 (2014)

    2014

  14. [20]

    A CMOS Spiking Neuron for Brain-Inspired Neural Networks With Resistive Synapses andIn SituLearning,

    X. Wu, V. Saxena, K. Zhu, and S. Balagopal, "A CMOS Spiking Neuron for Brain-Inspired Neural Networks With Resistive Synapses andIn SituLearning," IEEE Transactions on Circuits and Systems II: Express Briefs 62(11), 1088-1092 (2015)

    2015

  15. [21]

    Diannao: A small-footprint high- throughput accelerator for ubiquitous machine-learning,

    T. Chen, Z. Du, N. Sun, J. Wang, C. Wu, Y. Chen, and O. Temam, "Diannao: A small-footprint high- throughput accelerator for ubiquitous machine-learning," in ACM Sigplan Notices, (ACM, 2014), 269-284

    2014

  16. [22]

    Truenorth: Design and tool flow of a 65 mw 1 million neuron programmable neurosynaptic chip,

    F. Akopyan, J. Sawada, A. Cassidy, R. Alvarez-Icaza, J. Arthur, P. Merolla, N. Imam, Y. Nakamura, P. Datta, and G.-J. Nam, "Truenorth: Design and tool flow of a 65 mw 1 million neuron programmable neurosynaptic chip," IEEE Transactions on Computer-Aided Design of Integrated Ci...

    2015

  17. [23]

    Loihi: A neuromorphic manycore processor with on-chip learning,

    M. Davies, N. Srinivasa, T.-H. Lin, G. Chinya, Y. Cao, S. H. Choday, G. Dimou, P. Joshi, N. Imam, and S. Jain, "Loihi: A neuromorphic manycore processor with on-chip learning," IEEE Micro 38(1), 82-99 (2018)

    2018

  18. [24]

    Recent progress in semiconductor excitable lasers for photonic spike processing,

    P. R. Prucnal, B. J. Shastri, T. F. de Lima, M. A. Nahmias, and A. N. Tait, "Recent progress in semiconductor excitable lasers for photonic spike processing," Advances in Optics and Photonics 8(2), 228-299 (2016)

    2016

  19. [25]

    As alternative approaches to gradient - based methods, evolutionary algorithms are representative gradient free methods to optimize the weights of ANNs [41, 42]

    are trained by usi ng the BP and SGD algorithms . As alternative approaches to gradient - based methods, evolutionary algorithms are representative gradient free methods to optimize the weights of ANNs [41, 42]. Through the selection, crossover and mutation processes of a popu...

  20. [26]

    Optical computing,

    J. Touch, A.-H. Badawy, and V. J. Sorger, "Optical computing," Nanophotonics 6(3), 503-505 (2017)

    2017

  21. [27]

    A leaky integrate-and-fire laser neuron for ultrafast cognitive computing,

    M. A. Nahmias, B. J. Shastri, A. N. Tait, and P. R. Prucnal, "A leaky integrate-and-fire laser neuron for ultrafast cognitive computing," IEEE J. Sel. Top. Quantum Electron. 19(5), 1-12 (2013)

    2013

  22. [28]

    Multi-channel control for microring weight banks,

    A. N. Tait, T. F. De Lima, M. A. Nahmias, B. J. Shastri, and P. R. Prucnal, "Multi-channel control for microring weight banks," Opt. Express 24(8), 8895-8906 (2016)

    2016

  23. [29]

    Variance preserving initialization for training deep neuromorphic photonic networks with sinusoidal activations,

    N. Passalis, G. Mourgias-Alexandris, A. Tsakyridis, N. Pleros, and A. Tefas, "Variance preserving initialization for training deep neuromorphic photonic networks with sinusoidal activations," in ICASSP 2019- 2019 IEEE International Conference on Acoustics, Speech and Signal Pr...

    2019

  24. [30]

    Deep learning with coherent nanophotonic circuits,

    Y. Shen, N. C. Harris, S. Skirlo, M. Prabhu, T. Baehr-Jones, M. Hochberg, X. Sun, S. Zhao, H. Larochelle, D. Englund, and M. Soljačić, "Deep learning with coherent nanophotonic circuits," Nat. Photonics 11(7), 441-446 (2017)

    2017

  25. [32]

    Reinforcement learning in a large-scale photonic recurrent neural network,

    J. Bueno, S. Maktoobi, L. Froehly, I. Fischer, M. Jacquot, L. Larger, and D. Brunner, "Reinforcement learning in a large-scale photonic recurrent neural network," Optica 5(6), 756-760 (2018)

    2018

  26. [33]

    All-optical machine learning using diffractive deep neural networks,

    X. Lin, Y. Rivenson, N. T. Yardimci, M. Veli, Y. Luo, M. Jarrahi, and A. Ozcan, "All-optical machine learning using diffractive deep neural networks," Science 361(6406), 1004-1008 (2018)

    2018

  27. [34]

    Inverse design in nanophotonics,

    S. Molesky, Z. Lin, A. Y. Piggott, W. Jin, J. Vucković, and A. W. Rodriguez, "Inverse design in nanophotonics," Nat. Photonics 12(11), 659 (2018)

    2018

  28. [35]

    An all-optical neuron with sigmoid activation function,

    G. Mourgias-Alexandris, A. Tsakyridis, N. Passalis, A. Tefas, K. Vyrsokinos, and N. Pleros, "An all-optical neuron with sigmoid activation function," Opt. Express 27(7), 9620-9630 (2019)

    2019

  29. [37]

    Self-learning photonic signal processor with an optical neural network chip,

    H. Zhou, Y. Zhao, X. Wang, D. Gao, J. Dong, and X. Zhang, "Self-learning photonic signal processor with an optical neural network chip," arXiv preprint arXiv:1902.07318 (2019)

    2019 arXiv

  30. [38]

    Training of photonic neural networks through in situ backpropagation and gradient measurement,

    T. W. Hughes, M. Minkov, Y. Shi, and S. Fan, "Training of photonic neural networks through in situ backpropagation and gradient measurement," Optica 5(7), 864-871 (2018)

    2018

  31. [39]

    Binary particle swarm optimized 2× 2 power splitters in a standard foundry silicon photonic platform,

    J. C. Mak, C. Sideris, J. Jeong, A. Hajimiri, and J. K. Poon, "Binary particle swarm optimized 2× 2 power splitters in a standard foundry silicon photonic platform," Opt. Lett. 41(16), 3868-3871 (2016)

    2016

  32. [40]

    Silicon photonics circuit design: methods, tools and challenges,

    W. Bogaerts and L. Chrostowski, "Silicon photonics circuit design: methods, tools and challenges," Laser Photonics. Rev. 12(4), 1700237 (2018)

    2018

  33. [41]

    Efficient spectrum prediction and inverse design for plasmonic waveguide systems based on artificial neural networks,

    T. Zhang, J. Wang, Q. Liu, J. Zhou, J. Dai, X. Han, Y. Zhou, and K. Xu, "Efficient spectrum prediction and inverse design for plasmonic waveguide systems based on artificial neural networks," Photon. Res. 7(3), 368- 380 (2019)

    2019

  34. [42]

    Genetically optimized on-chip wideband ultracompact reflectors and Fabry–Perot cavities,

    Z. Yu, H. Cui, and X. Sun, "Genetically optimized on-chip wideband ultracompact reflectors and Fabry–Perot cavities," Photon. Res. 5(6), B15-B19 (2017)

    2017

  35. [43]

    Optimization for Gold Nanostructure-Based Surface Plasmon Biosensors Using a Microgenetic Algorithm,

    P.-H. Fu, S.-C. Lo, P.-C. Tsai, K.-L. Lee, and P.-K. Wei, "Optimization for Gold Nanostructure-Based Surface Plasmon Biosensors Using a Microgenetic Algorithm," ACS Photonics 5(6), 2320-2327 (2018)

    2018

  36. [44]

    Spiking neural networks for handwritten digit recognition—Supervised learning and network optimization,

    S. R. Kulkarni and B. Rajendran, "Spiking neural networks for handwritten digit recognition—Supervised learning and network optimization," Neural Networks 103, 118-127 (2018)

    2018

  37. [45]

    Designing neural networks through neuroevolution,

    K. O. Stanley, J. Clune, J. Lehman, and R. Miikkulainen, "Designing neural networks through neuroevolution," Nature Machine Intelligence 1(1), 24-35 (2019)

    2019

  38. [46]

    Deep neuroevolution: Genetic algorithms are a competitive alternative for training deep neural networks for reinforcement learning,

    F. P. Such, V. Madhavan, E. Conti, J. Lehman, K. O. Stanley, and J. Clune, "Deep neuroevolution: Genetic algorithms are a competitive alternative for training deep neural networks for reinforcement learning," arXiv preprint arXiv:1712.06567 (2017)

    2017 arXiv

  39. [47]

    Rainbow: Combining improvements in deep reinforcement learning,

    M. Hessel, J. Modayil, H. Van Hasselt, T. Schaul, G. Ostrovski, W. Dabney, D. Horgan, B. Piot, M. Azar, and D. Silver, "Rainbow: Combining improvements in deep reinforcement learning," in Thirty-Second AAAI Conference on Artificial Intelligence, (AAAI, 2018),

    2018

  40. [48]

    Deep learning in spiking neural networks,

    A. Tavanaei, M. Ghodrati, S. R. Kheradpisheh, T. Masquelier, and A. Maida, "Deep learning in spiking neural networks," Neural Networks 111, 47-63 (2018)

    2018

  41. [49]

    Optimal design for universal multiport interferometers,

    W. R. Clements, P. C. Humphreys, B. J. Metcalf, W. S. Kolthammer, and I. A. Walmsley, "Optimal design for universal multiport interferometers," Optica 3(12), 1460-1465 (2016)

    2016

  42. [50]

    Deep learning with spiking neurons: opportunities and challenges,

    M. Pfeiffer and T. Pfeil, "Deep learning with spiking neurons: opportunities and challenges," Frontiers in neuroscience 12(774), 1-18 (2018)

    2018

  43. [51]

    Nonlinear optics with 2D layered materials,

    A. Autere, H. Jussila, Y. Dai, Y. Wang, H. Lipsanen, and Z. Sun, "Nonlinear optics with 2D layered materials," Adv. Mater. 30(24), 1705963 (2018)

    2018

  44. [52]

    https://github.com/fancompute/neuroptica

  45. [53]

    Experimental realization of any discrete unitary operator,

    M. Reck, A. Zeilinger, H. J. Bernstein, and P. Bertani, "Experimental realization of any discrete unitary operator," Phys. Rev. Lett. 73(1), 58 (1994)

    1994

  46. [55]

    Feature selection based on hybridization of genetic algorithm and particle swarm optimization,

    P. Ghamisi and J. A. Benediktsson, "Feature selection based on hybridization of genetic algorithm and particle swarm optimization," IEEE Geosci. Remote Sens. Lett. 12(2), 309-313 (2015)

    2015

  47. [56]

    A study of cross-validation and bootstrap for accuracy estimation and model selection,

    R. Kohavi, "A study of cross-validation and bootstrap for accuracy estimation and model selection," in Ijcai, (Montreal, Canada, 1995), 1137-1145

    1995

  48. [57]

    Maximum certainty data partitioning,

    S. J. Roberts, R. Everson, and I. Rezek, "Maximum certainty data partitioning," Pattern Recognition 33(5), 833-839 (2000)

    2000

  49. [58]

    Automatic identification of digital modulation types,

    E. E. Azzouz and A. K. Nandi, "Automatic identification of digital modulation types," Signal Processing 47(1), 55-69 (1995)

    1995

  50. [200]

    This phenomenon is easy to explain because the large populations enhance the global searching ability of the evolution algorithms [36]

    are superior to that for the small population (N=50). This phenomenon is easy to explain because the large populations enhance the global searching ability of the evolution algorithms [36]. Although we can increase the population size of the PSO to achieve the lower MSE value ...

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.