arxiv: 2604.22270 · v1 · submitted 2026-04-24 · ⚛️ physics.optics

Recognition: unknown

Single-Shot Lensless Imaging with Physics Guided Genetic Programming

Ganesh M. Balasubramaniam , Xiao-Liu Chu , Radhika V. Nair , Matthew R. Foreman

Authors on Pith no claims yet

Pith reviewed 2026-05-08 10:52 UTC · model grok-4.3

classification ⚛️ physics.optics

keywords lensless imaginggenetic programmingsingle-shot reconstructionphase retrievalwave propagationmeta-optimizationoptical assay

0 comments

The pith

A genetically programmed iterative algorithm reconstructs complex objects from a single lensless intensity pattern by jointly estimating amplitude, phase, and detector distance.

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

The paper develops a single-shot lensless imaging method that recovers both amplitude and phase information using only one recorded intensity measurement. It does so by evolving an iterative reconstruction procedure through genetic programming, guided at each step by a physical wave-propagation model. The approach simultaneously optimizes the object amplitude, object phase, and the effective distance between object and detector. This matters for practical use because conventional lensless systems often need multiple exposures, precise calibration, or sample-specific retraining, which limits portability and robustness in diagnostics or monitoring. Experiments show the same evolved policy delivers high-quality results on resolution targets, micron beads, and biological cells across varied wavelengths and distances.

Core claim

We introduce a single-shot lensless imaging framework that reconstructs complex objects from only a single recorded intensity pattern using a genetically programmed iterative algorithm. Our method couples a wave-propagation model with an adaptive meta-optimisation strategy to jointly estimate the object amplitude, object phase, and effective object-detector distance. Experiments demonstrate high-fidelity recovery of amplitude objects, including a USAF target and 2 μm silicon beads on a glass slide, as well as a phase-dominant biological sample consisting of U2OS cells on a glass slide. Across multiple object types, wavelengths, and propagation distances, the same learned policy maintains高重建,

What carries the argument

Genetically programmed iterative algorithm that uses a wave-propagation model inside an adaptive meta-optimization loop to estimate object amplitude, phase, and effective propagation distance simultaneously.

If this is right

High-fidelity amplitude recovery for resolution targets and silicon beads, plus phase recovery for biological cells.
Integration into a wide-field β-amyloid bead assay for portable, single-shot measurements.
Consistent performance across object types, wavelengths, and distances using the identical learned policy.
Compact hardware becomes viable for point-of-care diagnostics and industrial monitoring where alignment tolerance and low cost matter.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The method could reduce reliance on extensive per-setup calibration in other computational imaging tasks that involve ill-posed inverse problems.
Evolutionary meta-optimization guided by physics models may serve as a lighter-weight alternative to deep-learning approaches in resource-constrained optical systems.
Testing the framework on dynamic or scattering samples would reveal whether the current generalization extends to more challenging biological or industrial scenes.
Adapting the same genetic-programming loop to different forward models might transfer the single-shot capability to modalities such as X-ray or acoustic imaging.

Load-bearing premise

The physical wave-propagation model used inside the optimization loop accurately describes light transport for the tested objects and distances.

What would settle it

Reconstruction quality dropping sharply on a new object type or at a propagation distance outside the training range, even when the same evolved policy is applied without retuning, would falsify the generalization claim.

read the original abstract

Lensless optical imaging eliminates the need for refractive optics, enabling compact and low-cost cameras with a large field-of-view, supporting point-of-care diagnostics and industrial monitoring. Practical deployments, however, remain constrained by ill-posed image reconstruction pipelines that require multiple measurements, careful calibration or object-specific training, thus limiting robustness and scalability. In this work, we introduce a single-shot lensless imaging framework that reconstructs complex objects from only a single recorded intensity pattern using a genetically programmed iterative algorithm. Our method couples a wave-propagation model with an adaptive meta-optimisation strategy to jointly estimate the object amplitude, object phase, and effective object-detector distance. Experiments demonstrate high-fidelity recovery of amplitude objects, including a USAF target and 2~$\mu$m silicon beads on a glass slide, as well as a phase-dominant biological sample consisting of U2OS cells on a glass slide. Across multiple object types, wavelengths, and propagation distances, the same learned policy maintains high reconstruction quality with minimal retuning, indicating strong out-of-distribution generalisation. As a practical demonstration, the framework is integrated with a $\beta$-amyloid-based optical digital bead assay under wide field-of-view acquisition. The resulting platform combines single-shot capture, compact hardware, and accurate reconstruction of complex fields, enabling rapid, portable assays in which throughput, alignment tolerance, and cost are critical.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Genetic programming for single-shot lensless reconstruction is a distinct idea but the generalization claims lack the quantitative details needed to evaluate them.

read the letter

The paper's core idea is to evolve an iterative reconstruction algorithm via genetic programming, guided by a wave-propagation model, so that a single intensity pattern yields joint estimates of object amplitude, phase, and distance. This is not a standard extension of existing phase-retrieval or learning methods, and the authors apply it to both amplitude and phase-dominant samples plus a practical bead assay for amyloid detection. That practical angle is useful and shows they are thinking about real deployment constraints like alignment tolerance and cost. The experiments span USAF targets, beads, cells, and the assay, which at least demonstrates breadth within the optics setting. The soft spots are more significant. The abstract asserts high-fidelity recovery and strong out-of-distribution generalization with only minimal retuning across object types, wavelengths, and distances, yet supplies no error metrics, no baseline comparisons, and no concrete description of what stayed fixed in the genetic programming setup or how retuning was measured. Without those, it is difficult to tell whether one fixed policy is doing the work or whether per-experiment adjustments are carrying the results. The stress-test note on unquantified retuning holds up on the available description. This is for researchers in computational optics who are exploring alternatives to conventional optimization or neural-network approaches for inverse problems. A reader already working on lensless hardware would find the framework worth examining for its different optimization strategy. I would send it for peer review so the implementation details and any unreported metrics can be checked directly.

Referee Report

3 major / 2 minor

Summary. The manuscript presents a single-shot lensless imaging method that employs a genetically programmed iterative algorithm coupling a wave-propagation forward model with an adaptive meta-optimization strategy. This jointly recovers object amplitude, object phase, and effective object-detector distance from one recorded intensity pattern. Experiments on amplitude targets (USAF resolution chart, 2 µm silicon beads), phase-dominant biological samples (U2OS cells), and a β-amyloid bead assay are reported to achieve high-fidelity reconstructions; the same learned policy is asserted to maintain performance across object types, wavelengths, and propagation distances with only minimal retuning, indicating out-of-distribution generalization.

Significance. If the generalization and fidelity claims hold under quantitative scrutiny, the approach could enable compact, calibration-light lensless systems for point-of-care diagnostics and industrial monitoring by eliminating the need for multiple exposures, object-specific training, or extensive alignment. The physics-guided genetic programming formulation offers a distinct route to solving the ill-posed inverse problem that may complement existing iterative phase-retrieval and learned-reconstruction techniques.

major comments (3)

[Abstract] Abstract: the assertions of 'high-fidelity recovery' and 'strong out-of-distribution generalisation' with 'minimal retuning' across object types, wavelengths, and distances are unsupported by any quantitative reconstruction metrics (PSNR, SSIM, phase/amplitude error), baseline comparisons, or error analysis, which directly undermines evaluation of the central performance claims.
[Experiments] Experiments section: no information is supplied on the exact hyper-parameters or population settings held constant across trials, nor is a quantitative measure of retuning (e.g., change in generations or mutation rate) or reconstruction metrics on deliberately held-out wavelengths/distances reported; this leaves open the possibility that per-experiment adaptation drives the results rather than a fixed policy.
[Method] Method section: details on how the genetic-programming meta-optimization avoids overfitting or incorporates post-hoc tuning are absent, making it impossible to verify that the learned policy remains independent of the target data as required for the out-of-distribution claim.

minor comments (2)

[Method] Notation for the wave-propagation operator and the genetic-programming fitness function should be introduced with explicit equations early in the method description to improve readability.
[Figures] Figure captions for the experimental reconstructions would benefit from inclusion of scale bars and a brief statement of the imaging wavelength and nominal propagation distance for each panel.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their constructive feedback, which has helped strengthen the quantitative rigor and clarity of our claims. We address each major comment below and have revised the manuscript accordingly.

read point-by-point responses

Referee: [Abstract] Abstract: the assertions of 'high-fidelity recovery' and 'strong out-of-distribution generalisation' with 'minimal retuning' across object types, wavelengths, and distances are unsupported by any quantitative reconstruction metrics (PSNR, SSIM, phase/amplitude error), baseline comparisons, or error analysis, which directly undermines evaluation of the central performance claims.

Authors: We agree that quantitative metrics are needed to support the abstract claims. In the revised manuscript we have added PSNR, SSIM, and mean phase/amplitude error values for all experiments, together with comparisons against the Gerchberg-Saxton algorithm and a standard iterative phase-retrieval baseline. Standard-deviation error bars across repeated runs are now reported in the Experiments section. revision: yes
Referee: [Experiments] Experiments section: no information is supplied on the exact hyper-parameters or population settings held constant across trials, nor is a quantitative measure of retuning (e.g., change in generations or mutation rate) or reconstruction metrics on deliberately held-out wavelengths/distances reported; this leaves open the possibility that per-experiment adaptation drives the results rather than a fixed policy.

Authors: We have added a table in the revised Experiments section listing all hyper-parameters and population settings (population size, mutation rate, number of generations, crossover probability) that remained fixed across trials. We also quantify retuning as the fractional change in generations (typically <10 %) and report PSNR/SSIM on deliberately held-out wavelengths and propagation distances that were excluded from policy learning. revision: yes
Referee: [Method] Method section: details on how the genetic-programming meta-optimization avoids overfitting or incorporates post-hoc tuning are absent, making it impossible to verify that the learned policy remains independent of the target data as required for the out-of-distribution claim.

Authors: The fitness function is defined exclusively by the physics-based wave-propagation model, which acts as a hard constraint and prevents data-specific overfitting. We have expanded the Methods section to describe the post-hoc tuning procedure: only a small set of fitness-weight coefficients are adjusted on an independent validation set; the core policy parameters are never updated on target data. This detail now explicitly supports the out-of-distribution claim. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper's core method couples an external wave-propagation model (standard physics) with genetic programming for adaptive optimization to jointly recover amplitude, phase, and distance from single-shot intensity data. No derivation step reduces by construction to its own inputs, fitted parameters, or self-citations. The generalization claim across object types, wavelengths, and distances rests on reported experiments rather than tautological renaming or self-referential definitions. The provided abstract and context contain no load-bearing self-citation chains or ansatz smuggling that would elevate the circularity score.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the standard wave-propagation model being an accurate forward operator and on genetic programming being able to discover a generalizable policy; no new physical entities are postulated and no free parameters are explicitly fitted in the abstract description.

axioms (1)

domain assumption Wave propagation model accurately describes light transport from object to detector
The method explicitly couples this model to the genetic programming loop for reconstruction.

pith-pipeline@v0.9.0 · 5558 in / 1185 out tokens · 31834 ms · 2026-05-08T10:52:51.549719+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

38 extracted references · 23 canonical work pages

[1]

Automatic mitigation of dynamic atmospheric turbulence using optical phase conjugation for coher- ent free-space optical communications

Boominathan, V., Robinson, J.T., Waller, L., Veeraraghavan, A.: Recent advances in lensless imaging. Optica9(1), 1–16 (2022) https://doi.org/10.1364/OPTICA. 431361

work page doi:10.1364/optica 2022
[2]

Annual Review of Biomedical Engineering18, 77–102 (2016) https://doi.org/10.1146/ annurev-bioeng-092515-010849

Ozcan, A., McLeod, E.: Lensless imaging and sensing. Annual Review of Biomedical Engineering18, 77–102 (2016) https://doi.org/10.1146/ annurev-bioeng-092515-010849

2016
[3]

Applied Physics B130(9), 166 (2024)

Rosen, J., Alford, S., Allan, B., Anand, V., Arnon, S., Arockiaraj, F.G., Art, J., Bai, B., Balasubramaniam, G.M., Birnbaum, T.,et al.: Roadmap on computa- tional methods in optical imaging and holography. Applied Physics B130(9), 166 (2024)

2024
[4]

Laser & Photonics Reviews18(10), 2400197 (2024) 21

Potter, C.J., Xiong, Z., McLeod, E.: Clinical and biomedical applications of lensless holographic microscopy. Laser & Photonics Reviews18(10), 2400197 (2024) 21

2024
[5]

Lab on a Chip9(6), 777–787 (2009) https://doi.org/10.1039/B813943A

Seo, S., Su, T.-W., Tseng, D.K., Erlinger, A., Ozcan, A.: Lensfree holographic imaging for on-chip cytometry and diagnostics. Lab on a Chip9(6), 777–787 (2009) https://doi.org/10.1039/B813943A

work page doi:10.1039/b813943a 2009
[6]

Proceedings of the National Academy of Sciences of the United States of America112(18), 5613–5618 (2015) https: //doi.org/10.1073/pnas.1501815112

Im, H., Castro, C.M., Shao, H., Liong, M., Song, J., Pathania, D., Fexon, L., Min, C., Avila-Wallace, M., Zurkiya, O., Rho, J., Magaoay, B., Tambouret, R.H., Pivovarov, M., Weissleder, R., Lee, H.: Digital diffraction analysis enables low- cost molecular diagnostics on a smartphone. Proceedings of the National Academy of Sciences of the United States of A...

work page doi:10.1073/pnas.1501815112 2015
[7]

IEEE Signal Processing Magazine32(3), 87–109 (2015) https://doi.org/10.1109/MSP.2015.2398954

Shechtman, Y., Eldar, Y.C., Cohen, O., Chapman, H.N., Miao, J., Segev, M.: Phase retrieval with application to optical imaging. IEEE Signal Processing Magazine32(3), 87–109 (2015) https://doi.org/10.1109/MSP.2015.2398954

work page doi:10.1109/msp.2015.2398954 2015
[8]

Optics and Lasers in Engineering172, 107878 (2024) https://doi.org/10.1016/j.optlaseng.2023

Arcab, P., Rogalski, M., Trusiak, M.: Single-shot experimental-numerical twin- image removal in lensless digital holographic microscopy. Optics and Lasers in Engineering172, 107878 (2024) https://doi.org/10.1016/j.optlaseng.2023. 107878

work page doi:10.1016/j.optlaseng.2023 2024
[9]

Optik35, 237–246 (1972)

Gerchberg, R.W., Saxton, W.O.: A practical algorithm for the determination of phase from image and diffraction plane pictures. Optik35, 237–246 (1972)

1972
[10]

Applied Optics21(15), 2758–2769 (1982) https://doi.org/10.1364/AO.21.002758

Fienup, J.R.: Phase retrieval algorithms: A comparison. Applied Optics21(15), 2758–2769 (1982) https://doi.org/10.1364/AO.21.002758

work page doi:10.1364/ao.21.002758 1982
[11]

Optics Express20(3), 3129–3143 (2012) https://doi.org/10.1364/OE.20.003129

Greenbaum, A., Ozcan, A.: Maskless imaging of dense samples using pixel super- resolution based multi-height lensfree on-chip microscopy. Optics Express20(3), 3129–3143 (2012) https://doi.org/10.1364/OE.20.003129

work page doi:10.1364/oe.20.003129 2012
[12]

Ultramicroscopy109(10), 1256–1262 (2009) https://doi.org/10.1016/j.ultramic.2009.05.012

Maiden, A.M., Rodenburg, J.M.: An improved ptychographical phase retrieval algorithm for diffractive imaging. Ultramicroscopy109(10), 1256–1262 (2009) https://doi.org/10.1016/j.ultramic.2009.05.012

work page doi:10.1016/j.ultramic.2009.05.012 2009
[13]

Nature Photonics7, 739–745 (2013) https://doi.org/10.1038/ nphoton.2013.187

Zheng, G., Horstmeyer, R., Yang, C.: Wide-field, high-resolution fourier ptycho- graphic microscopy. Nature Photonics7, 739–745 (2013) https://doi.org/10.1038/ nphoton.2013.187

2013
[14]

IEEE Transactions on Computational Imaging3(3), 384–397 (2017) https://doi.org/ 10.1109/TCI.2016.2593662

Asif, M.S., Ayremlou, A., Sankaranarayanan, A.C., Veeraraghavan, A., Baraniuk, R.: Flatcam: Thin, lensless cameras using coded aperture and computation. IEEE Transactions on Computational Imaging3(3), 384–397 (2017) https://doi.org/ 10.1109/TCI.2016.2593662

work page doi:10.1109/tci.2016.2593662 2017
[15]

Optica5(1), 1–9 (2018) https: //doi.org/10.1364/OPTICA.5.000001 22

Antipa, N., Kuo, G., Heckel, R., Mildenhall, B., Bostan, E., Ng, R., Waller, L.: Diffusercam: Lensless single-exposure 3d imaging. Optica5(1), 1–9 (2018) https: //doi.org/10.1364/OPTICA.5.000001 22

work page doi:10.1364/optica.5.000001 2018
[16]

1992 , issn =

Rudin, L.I., Osher, S., Fatemi, E.: Nonlinear total variation based noise removal algorithms. Physica D: Nonlinear Phenomena60(1–4), 259–268 (1992) https:// doi.org/10.1016/0167-2789(92)90242-F

work page doi:10.1016/0167-2789(92)90242-f 1992
[17]

Scientific Reports6, 37862 (2016) https://doi.org/10.1038/srep37862

Rivenson, Y., Weiss, A., Stern, A., Zalevsky, Z., Ozcan, A.: Sparsity-based multi- height phase recovery in holographic microscopy. Scientific Reports6, 37862 (2016) https://doi.org/10.1038/srep37862

work page doi:10.1038/srep37862 2016
[18]

Light: Science & Applications7, 17141 (2018) https://doi.org/10

Rivenson, Y., Wu, Y., Zhang, Y., Weiss, Z., G¨ unaydin, H., Lin, X., Ozcan, A.: Phase recovery and holographic image reconstruction using deep learning in neu- ral networks. Light: Science & Applications7, 17141 (2018) https://doi.org/10. 1038/lsa.2017.141

2018
[19]

Optica5(6), 704–710 (2018) https://doi

Wu, Y., Rivenson, Y., Wang, H., Luo, Y., Ben-David, E., Stern, A., Ozcan, A.: Extended depth of field in holographic image reconstruction using deep learning based autofocusing and phase recovery. Optica5(6), 704–710 (2018) https://doi. org/10.1364/OPTICA.5.000704

work page doi:10.1364/optica.5.000704 2018
[20]

Optics Express26(18), 22603–22614 (2018) https://doi.org/10.1364/OE.26.022603

Wang, H., Lyu, M., Situ, G.: eholonet: a learning based end to end approach for in line digital holographic reconstruction. Optics Express26(18), 22603–22614 (2018) https://doi.org/10.1364/OE.26.022603

work page doi:10.1364/oe.26.022603 2018
[21]

IEEE Access8, 202648–202659 (2020) https://doi.org/10.1109/ACCESS.2020.3036380

Li, H., Chen, X., Chi, Z., Mann, C., Razi, A.: Deep dih: Single shot digital in line holography reconstruction by deep learning. IEEE Access8, 202648–202659 (2020) https://doi.org/10.1109/ACCESS.2020.3036380

work page doi:10.1109/access.2020.3036380 2020
[22]

Optica7(6), 559–562 (2020) https://doi.org/10.1364/OPTICA.387381

Bostan, E., Heckel, R., Chen, M., Kellman, M., Waller, L.: Self calibrating phase microscopy with an untrained deep neural network. Optica7(6), 559–562 (2020) https://doi.org/10.1364/OPTICA.387381

work page doi:10.1364/optica.387381 2020
[23]

Phase imaging with an untrained neural network

Wang, F., Bian, Y., Wang, H., Lyu, M., Pedrini, G., Osten, W., Barbastathis, G., Situ, G.: Phase imaging with an untrained neural network. Light: Science & Applications9, 77 (2020) https://doi.org/10.1038/s41377-020-0302-3

work page doi:10.1038/s41377-020-0302-3 2020
[24]

Optics Express32(6), 10444–10460 (2024) https://doi.org/10

Zhang, Y., Liu, X., Lam, E.Y.: Single shot inline holography using a physics aware diffusion model. Optics Express32(6), 10444–10460 (2024) https://doi.org/10. 1364/OE.517233

2024
[25]

Scientific Reports13, 11105 (2023) https://doi.org/10.1038/ s41598-023-37810-w

Manisha, Mandal, A.C.,et al.: Randomness assisted in line holography with deep learning. Scientific Reports13, 11105 (2023) https://doi.org/10.1038/ s41598-023-37810-w

2023
[26]

Nature Communications16, 4840 (2025) https://doi.org/10.1038/s41467-025-60200-x 23

Kim, J., Kim, Y., Lee, H.S., Seo, E., Lee, S.J.: Single-shot reconstruction of three-dimensional morphology of biological cells in digital holographic microscopy using a physics-driven neural network. Nature Communications16, 4840 (2025) https://doi.org/10.1038/s41467-025-60200-x 23

work page doi:10.1038/s41467-025-60200-x 2025
[27]

MIT Press, Cambridge, MA (1992)

Koza, J.R.: Genetic Programming: On the Programming of Computers by Means of Natural Selection. MIT Press, Cambridge, MA (1992)

1992
[28]

Lulu.com, Freely available at http://www.gp-field-guide.org.uk (2008)

Poli, R., Langdon, W.B., McPhee, N.F., Koza, J.R.: A Field Guide to Genetic Programming. Lulu.com, Freely available at http://www.gp-field-guide.org.uk (2008)

2008
[29]

Springer, Berlin (2015)

Eiben, A.E., Smith, J.E.: Introduction to Evolutionary Computing, 2nd edn. Springer, Berlin (2015). https://doi.org/10.1007/978-3-662-44874-8

work page doi:10.1007/978-3-662-44874-8 2015
[30]

Computational Intelligence37(4), 1745–1778 (2021) https://doi.org/10.1111/coin.12459

Khan, A., Qureshi, A.S., Wahab, N., Hussain, M., Hamza, M.Y.: A recent survey on the applications of genetic programming in image processing. Computational Intelligence37(4), 1745–1778 (2021) https://doi.org/10.1111/coin.12459

work page doi:10.1111/coin.12459 2021
[31]

In preparation

Nair, R.V., Balasubramaniam, G.M., Chu, X.-L., Parikh, A., Foreman, M.R.: Quantitative Digital Bead Profiling of EPPS Mediated Amyloid-βProtofibril Disassembly. In preparation
[32]

Nature Communications6, 8997 (2015) https://doi.org/10.1038/ncomms9997

Kim, H.Y., Kim, H.V., Jo, S., Lee, C.J., Choi, S.Y., Kim, D.J., Kim, Y.: Epps rescues hippocampus-dependent cognitive deficits in app/ps1 mice by disaggrega- tion of amyloid-βoligomers and plaques. Nature Communications6, 8997 (2015) https://doi.org/10.1038/ncomms9997

work page doi:10.1038/ncomms9997 2015
[33]

bioRxiv (2026) https://doi.org/10.64898/2026.02.22.706957 https://www.biorxiv.org/content/early/2026/02/23/2026.02.22.706957.full.pdf

Nair, R.V., Tran, B.N., Parikh, A.N., Foreman, M.R.: Small molecule ensembles reshape amyloid aggregation land- scapes. bioRxiv (2026) https://doi.org/10.64898/2026.02.22.706957 https://www.biorxiv.org/content/early/2026/02/23/2026.02.22.706957.full.pdf

work page doi:10.64898/2026.02.22.706957 2026
[34]

Roberts and Company Publishers, Englewood, CO (2005)

Goodman, J.W.: Introduction to Fourier Optics, 3rd edn. Roberts and Company Publishers, Englewood, CO (2005). ISBN-10: 0974707724; xvi+491 pp

2005
[35]

SPIE Tutorial Texts, vol

Voelz, D.G.: Computational Fourier Optics: A MATLAB Tutorial. SPIE Tutorial Texts, vol. TT89. SPIE Press, Bellingham, WA (2011). https://doi.org/10.1117/ 3.858456 . PDF eISBN: 978-0-8194-8205-1

2011
[36]

The Annals of Mathemat- ical Statistics35(1), 73–101 (1964) https://doi.org/10.1214/aoms/1177703732

Huber, P.J.: Robust estimation of a location parameter. The Annals of Mathemat- ical Statistics35(1), 73–101 (1964) https://doi.org/10.1214/aoms/1177703732

work page doi:10.1214/aoms/1177703732 1964
[37]

In: Interna- tional Conference on Learning Representations (2015)

Kingma, D.P., Ba, J.: Adam, a method for stochastic optimization. In: Interna- tional Conference on Learning Representations (2015)

2015
[38]

Wiley, Chichester (2001) 24

Deb, K.: Multi-Objective Optimization Using Evolutionary Algorithms. Wiley, Chichester (2001) 24

2001