Pith. sign in

REVIEW 4 major objections 3 minor 1 cited by

Mass Loss and Subsequent Thermal Evolution of Surviving Helium White Dwarfs Shocked by Thermonuclear Supernovae

T0 review · 4 major / 3 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read The paper argues that a single pressure ratio controls how much mass a surviving helium white dwarf loses in a Type Ia supernova, and that this ratio sets the structure of the resulting hypervelocity white dwarfs.

desk verdict Wrong PDF uploaded: the body is a computer-vision paper, not the advertised helium WD study; desk reject and ask the authors to resubmit. read the letter →

arxiv 2508.12529 v1 pith:VVYNB2BW submitted 2025-08-17 astro-ph.SR astro-ph.HE

Tin Long Sunny Wong , Lars Bildsten This is my paper · ORCID
classification astro-ph.SRastro-ph.HE
keywords TypeIasupernovaeheliumwhitedwarfshypervelocitymasslosssupernovaejectainteractiondoubledwarfbinariesshockheatingstellarevolution
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 argues that a surviving helium white dwarf's fractional mass loss in a Type Ia supernova is controlled by a single number: the ratio of the supernova ejecta's ram pressure to the white dwarf's volume-averaged internal pressure. Using three-dimensional hydrodynamic simulations with a realistic degenerate-matter equation of state, the authors find a tight relation between this pressure ratio and the fractional mass loss, so the stripping in any given double-white-dwarf system can be predicted without a new simulation. Lower-mass, larger-radius helium white dwarfs are hit hardest; the most extreme survivors are left as very inflated objects with masses of about \(0.02\,M_\odot\). The subsequent thermal evolution of the shock-heated survivors matches the inflated radii of four observed hypervelocity white dwarfs, and the predicted mass-loss rates imply the ejected helium should be detectable as nebular emission lines. If right, the paper gives a practical way to tie observed hypervelocity white dwarfs and supernova remnants back to the orbital and structural parameters of the pre-explosion binary.

What carries the argument

The load-bearing quantity is the dimensionless ratio \(\mathcal{R}=P_{\rm ram}/\bar{P}\), where \(P_{\rm ram}\) is the ram pressure of the supernova ejecta at the companion and \(\bar{P}\) is the volume-averaged internal pressure of the donor white dwarf. The paper shows that fractional mass loss collapses onto a tight relation with this single ratio, turning it into a predictor for systems beyond the simulated grid. The supporting machinery is a three-dimensional hydrodynamics treatment with an accurate equation of state for degenerate helium, which captures both the stripping and the shock-deposited entropy, together with a post-shock thermal-evolution calculation that converts the shocked

What would settle it

Detecting a hot, inflated helium white dwarf at the center of SN 0509-67.5 or SN 1006 would contradict the paper's exclusion of any surviving helium white dwarf there; measuring a hypervelocity white dwarf whose mass, radius, and velocity place it far from the predicted pressure-ratio thermal track would falsify the scaling.

Watch

Extended reading notes

Core claim

The central claim: a Roche lobe-filling helium white dwarf struck by Type Ia ejecta loses a mass fraction set by \(\mathcal{R}=P_{\rm ram}/\bar{P}\), ejecta ram pressure over donor volume-averaged pressure. Three-dimensional simulations of \(0.08\)–\(0.45\,M_\odot\) donors find lower-mass, larger-radius donors lose the most; the extreme survivor is an inflated \(\sim0.02\,M_\odot\) object. Helium loss \(0.005\)–\(0.06\,M_\odot\) at \(1{,}000\)–\(4{,}000\,\mathrm{km\,$s^{{-1}}$}\) should power nebular lines, and final velocity is essentially orbital (\(\lesssim1{,}600\,\mathrm{km\,$s^{{-1}}$}\)). Post-shock thermal models match four inflated hypervelocity white dwarfs; they exclude \(\gtrsim0.3\,M_\o

Load-bearing premise

The simulations assume the helium white dwarf donor is filling its Roche lobe at the moment of the explosion; if the donor is detached, the ejecta interaction geometry, the mass loss, and the post-shock radii would all differ.

Editorial extensions

If this is right

  • Stripped masses of surviving helium white dwarfs can be predicted for any double-white-dwarf Type Ia system from the ejecta and donor properties, not just the simulated grid.
  • The inflated radii of observed hypervelocity white dwarfs can be matched to particular pre-explosion donor masses and orbital configurations through their post-shock thermal tracks.
  • Ejected helium masses of \(0.005\)–\(0.06\,M_\odot\) at \(1{,}000\)–\(4{,}000\,\mathrm{km\,s^{-1}}\) should produce detectable nebular-phase helium emission lines in nearby Type Ia supernovae.
  • The absence of a surviving \(\gtrsim0.3\,M_\odot\) helium white dwarf in SN 1972E and SN 2011fe, and of essentially any helium white dwarf in SN 0509-67.5 and SN 1006, becomes a testable constraint on those supernova progenitors.
  • Because the final velocity is essentially the orbital velocity, measured space velocities of hypervelocity white dwarfs directly probe the pre-explosion binary orbit rather than an explosion kick.

Reading between the lines

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

  • The pressure-ratio scaling may generalize beyond helium white dwarfs to carbon-oxygen or even non-degenerate companions, provided their volume-averaged pressure is used; a small grid of such simulations would test this.
  • The predicted nebular helium emission lines offer a direct search target: late-time spectra of nearby Type Ia remnants should show helium in \(10^3\,\mathrm{km\,s^{-1}}\) wide lines if a helium white dwarf companion was stripped.
  • If a hypervelocity white dwarf is found with a radius that disagrees with the predicted thermal track, the likely explanation is that the donor did not fill its Roche lobe or the ejecta were asymmetric, making the comparison a geometric diagnostic rather than a failure of the relation.
  • The relation could be calibrated empirically: combined mass, radius, and space-velocity measurements of the inflated hypervelocity white dwarfs constrain the pre-explosion orbital separation for each object.
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 / 3 minor

Summary. The abstract of arXiv:2508.12529 (astro-ph.SR) describes a study of SN Ia ejecta interaction with Roche-lobe-filling helium WD companions, using Athena++ 3D hydrodynamics with a degenerate He equation of state and MESA thermal evolution. It claims a tight mass-loss relation, substantial mass loss, inflated surviving companions, agreement with hypervelocity WDs, and constraints on surviving companions in historical SNRs. However, the submitted full text is the paper 'MuSACo: Multimodal Subject-Specific Selection and Adaptation for Expression Recognition with Co-Training' (a computer-vision paper). No astrophysical methods, simulations, equations, tables, figures, or observational comparisons appear in the manuscript. Consequently, every astrophysical claim in the abstract is unsupported by the submitted text.

Significance. If the claimed results were backed by the described simulations, they would be of value: a predictive mass-loss relation for shocked WDs, post-shock radii for comparison to Gaia hypervelocity WDs, and constraints on SN Ia companion survival. However, significance cannot be assessed because the manuscript contains none of the supporting material. The paper offers no reproducible code, no machine-checked proofs, no parameter-free derivations, and no falsifiable predictions that can be checked from the submitted text.

major comments (4)
  1. [Full text] The entire body of the submission (Sections 1-13, all tables, all algorithms) is a computer-vision paper on multimodal expression recognition. It contains no mention of Athena++, MESA, helium white dwarfs, thermonuclear supernovae, hypervelocity stars, or any of the objects named in the abstract. The central claims are therefore not merely insufficiently supported; the manuscript does not contain the described study.
  2. [Abstract] The claimed tight relation between fractional mass loss and the ratio of ejecta ram pressure to donor volume-averaged pressure is never defined, derived, or evaluated. No equation, simulation grid, initial conditions, EOS description, or convergence/error analysis is provided. This is a load-bearing missing element because the entire predictive utility of the paper rests on this relation.
  3. [Abstract (observational comparisons)] The claimed agreement with D6-2, J0546+0836, J1332-3541, and SDSS J1637+3631, and the ruling out of surviving He WDs in SN1972E, SN2011fe, 0509-67.5, and SN1006, are presented without any data, fitting procedure, or uncertainty quantification. These conclusions cannot be checked or reproduced from the manuscript.
  4. [Abstract (assumption)] The assumption that the He WD companion is Roche lobe-filling at the moment of explosion is stated in the abstract but nowhere justified or tested. Since the ejecta interaction geometry, mass loss, and post-shock radii all depend on the donor's configuration at explosion, this assumption is load-bearing and remains entirely unexamined in the submitted text.
minor comments (3)
  1. [Title/metadata] The arXiv title and abstract (astro-ph.SR) do not match the full text (cs.CV). The manuscript should either contain the described astrophysics study or be withdrawn/corrected.
  2. [References] All references in the full text are to computer-vision and machine-learning literature; no astrophysical references are present, so the context (e.g., previous SN Ia companion studies, Gaia hypervelocity WD discoveries) is missing.
  3. [Reproducibility] No data availability statement, simulation scripts, or MESA/Athena++ input files are provided for the claimed astrophysical results.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the manuscript body is an unrelated computer-vision paper, so the claimed astrophysical derivation is absent and cannot be evaluated for circularity.

full rationale

The submitted full text is the paper 'MuSACo: Multimodal Subject-Specific Selection and Adaptation for Expression Recognition with Co-Training' by Zeeshan et al., a computer vision manuscript with no equations, simulations, or data related to the abstract's claims about shocked helium white dwarfs, Athena++ simulations, or MESA modeling. Because the claimed derivation chain is entirely absent from the provided text, there is no specific equation, self-citation, or fitted parameter that can be quoted to exhibit a reduction of a prediction to its inputs. Under the hard rule that circularity must be demonstrated with direct evidence, no circular step can be identified. The mismatch between abstract and body is a serious correctness/completeness defect, but it is not a self-consistency circularity. Accordingly, the honest finding is no detectable circularity in the submitted material.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

No new particles, forces, or entities are introduced in the abstract. The study uses known physical models and computational tools.

free parameters (1)
  • All simulation input parameters (ejecta mass, energy, density profile; binary separation; EOS details)
    The abstract does not enumerate which parameters are varied or fitted. Without the full text the free-parameter count cannot be audited.
assumptions (3)
  • ad hoc to paper The helium WD donor is Roche lobe-filling at the moment of explosion.
    The abstract explicitly restricts the study to Roche lobe-filling 0.08 to 0.45 solar mass helium WD companions. If the donor is detached, the interaction geometry changes.
  • domain assumption The Athena++ 3D hydrodynamics with the adopted degenerate helium EOS accurately represents the ejecta-donor interaction and subsequent mass loss.
    All quantitative claims about mass loss depend on the numerical fidelity of the simulations and the EOS, neither of which can be verified from the abstract.
  • domain assumption The observed hypervelocity WDs (D6-2, J0546+0836, J1332-3541, SDSS J1637+3631) are shock-heated surviving companions of SNe Ia as modeled.
    The MESA comparisons and remnant exclusions rely on these identifications and on the assumption that the modeled initial conditions apply to those systems.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Mass Loss and Subsequent Thermal Evolution of Surviving Helium White Dwarfs Shocked by Thermonuclear Supernovae." pith.science (2026). https://pith.science/paper/VVYNB2BW

@misc{pith2026250812529,
  author       = {Pith},
  title        = {Pith review of: Mass Loss and Subsequent Thermal Evolution of Surviving Helium White Dwarfs Shocked by Thermonuclear Supernovae},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VVYNB2BW}},
  note         = {Machine review of arXiv:2508.12529}
}
abstract

Following a type Ia supernova (SN Ia) in a double white dwarf (WD) binary, a surviving WD companion leaves at its orbital velocity $\approx 1$,000 - 3,000 km/s. The Gaia mission has discovered seven such hypervelocity WDs with inflated radii indicative of shock heating by SN ejecta. We study the interaction between SN ejecta and Roche lobe-filling 0.08 - 0.45 $M_{\odot}$ helium WD companions using three-dimensional hydrodynamical simulations with Athena++. Given the importance of the later thermal evolution, we include an accurate equation-of-state for the degenerate helium WD donor. We show that a lower-mass, larger-radius WD companion is more strongly impacted by SN ejecta and undergoes substantial mass loss. We find a tight relation between the fractional mass loss and the ratio between the ejecta ram pressure and donor volume-averaged pressure, which can be used for predicting mass loss in other systems. In the most extreme case, the companion becomes a very inflated $\approx0.02\,M_{\odot}$ object. We find helium mass loss $\approx 0.005 - 0.06\,M_{\odot}$ with velocities $\approx $ 1,000$-$4,000 km/s, which may lead to emission lines in the nebular phase. The surviving helium WD receives a kick velocity, but its final velocity is essentially determined by its orbital velocity $\lesssim$ 1,600 km/s. We model the post-explosion evolution of the shock-heated companions using MESA, and find reasonable agreement with the hypervelocity stars D6-2, J0546+0836, J1332-3541 & SDSS J1637+3631. A surviving $\gtrsim 0.3\,M_{\odot}$ helium WD can be ruled out in SN1972E & SN2011fe, and any surviving helium WD is likely ruled out in SN remnants 0509-67.5 & SN1006.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. V407 Vul: a triple star system with an AM CVn detectable by gravitational wave observatories

    astro-ph.SR 2026-07 accept novelty 7.0 of 10

    V407 Vul is a hierarchical triple: a 569-s AM CVn white-dwarf binary bound to a G-type tertiary at ~120 AU, the first 'verification triple' for LISA.

Reference graph

Works this paper leans on

51 extracted references · 48 canonical work pages · cited by 1 Pith paper

  1. [1]

    Privileged knowledge distillation for dimen- sional emotion recognition in the wild

    Muhammad Haseeb Aslam, Muhammad Osama Zeeshan, Marco Pedersoli, Alessandro L Koerich, Simon Bacon, and Eric Granger. Privileged knowledge distillation for dimen- sional emotion recognition in the wild. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 3338–3347, 2023. 1, 2, 3, 5

    work page 2023

  2. [2]

    Distilling privi- leged multimodal information for expression recognition us- ing optimal transport

    Muhammad Haseeb Aslam, Muhammad Osama Zeeshan, Soufiane Belharbi, Marco Pedersoli, Alessandro Lameiras Koerich, Simon Bacon, and Eric Granger. Distilling privi- leged multimodal information for expression recognition us- ing optimal transport. In18th International Conference on Automatic Face and Gesture Recognition (FG), pages 1–10,

    work page

  3. [3]

    Combining labeled and un- labeled data with co-training

    Avrim Blum and Tom Mitchell. Combining labeled and un- labeled data with co-training. InProceedings of the eleventh annual conference on Computational learning theory, pages 92–100, 1998. 3

    work page 1998

  4. [4]

    Stressid: a multimodal dataset for stress iden- tification.Advances in Neural Information Processing Sys- tems, 36:29798–29811, 2023

    Hava Chaptoukaev, Valeriya Strizhkova, Michele Panariello, Bianca Dalpaos, Aglind Reka, Valeria Manera, Susanne Th¨ummler, Esma Ismailova, Massimiliano Todisco, Maria A Zuluaga, et al. Stressid: a multimodal dataset for stress iden- tification.Advances in Neural Information Processing Sys- tems, 36:29798–29811, 2023. 6, 10

    work page 2023

  5. [5]

    Co- training for domain adaptation.Advances in neural informa- tion processing systems, 24, 2011

    Minmin Chen, Kilian Q Weinberger, and John Blitzer. Co- training for domain adaptation.Advances in neural informa- tion processing systems, 24, 2011. 3

    work page 2011

  6. [6]

    Robust target training for multi-source domain adaptation

    Zhongying Deng, Da Li, Yi-Zhe Song, and Tao Xiang. Robust target training for multi-source domain adaptation. CoRR, 2022. 3, 5, 6

    work page 2022

  7. [7]

    Subject-wise F1 scores of the BioVid dataset using the ViT backbone across 10 target subjects

    Andrius Dzedzickis, Art ¯uras Kaklauskas, and Vytautas Settings Methods Target Subjects Sub-1 Sub-2 Sub-3 Sub-4 Sub-5 Sub-6 Sub-7 Sub-8 Sub-9 Sub-10 Avg F1 Lower Bound Visual-only Physio-only Fusion 35.6 31.0 23.9 27.5 21.9 27.1 22.5 23.1 22.8 12.5 24.5 22.6 19.7 27.9 33.5 19.6 23.6 9.5 35.3 23.9 32.9 15.6 27.3 34.6 14.9 28.2 29.1 23.2 19.1 26.1 22.6 25.0...

    work page 2020

  8. [8]

    Domain-adversarial training of neural networks.Journal of machine learning research, 17(59):1–35, 2016

    Yaroslav Ganin, Evgeniya Ustinova, Hana Ajakan, Pas- cal Germain, Hugo Larochelle, Franc ¸ois Laviolette, Mario March, and Victor Lempitsky. Domain-adversarial training of neural networks.Journal of machine learning research, 17(59):1–35, 2016. 6, 7, 15

    work page 2016

Show all 51 references
  1. [9]

    Co-training for unsupervised domain adaptation of semantic segmentation models.Sensors, 23(2):621, 2023

    Jose L G ´omez, Gabriel Villalonga, and Antonio M L ´opez. Co-training for unsupervised domain adaptation of semantic segmentation models.Sensors, 23(2):621, 2023. 3

    2023

  2. [10]

    Bah dataset for ambivalence/hesitancy recognition in videos for behavioural change.arXiv preprint arXiv:2505.19328, 2025

    Manuela Gonz ´alez-Gonz´alez, Soufiane Belharbi, Muham- mad Osama Zeeshan, Masoumeh Sharafi, Muham- mad Haseeb Aslam, Marco Pedersoli, Alessandro Lameiras Koerich, Simon L Bacon, and Eric Granger. Bah dataset for ambivalence/hesitancy recognition in videos for behavioural chan...

    2025 arXiv

  3. [11]

    A kernel two-sample test.The Journal of Machine Learning Research, 13(1):723– 773, 2012

    Arthur Gretton, Karsten M Borgwardt, Malte J Rasch, Bern- hard Sch¨olkopf, and Alexander Smola. A kernel two-sample test.The Journal of Machine Learning Research, 13(1):723– 773, 2012. 1

    2012

  4. [12]

    Source-free domain adaptation of weakly- supervised object localization models for histology

    Alexis Guichemerre, Soufiane Belharbi, Tsiry Mayet, Sha- keeb Murtaza, Pourya Shamsolmoali, Luke McCaffrey, and Eric Granger. Source-free domain adaptation of weakly- supervised object localization models for histology. InPro- ceedings of the IEEE/CVF Conference on Computer Vi...

    2024

  5. [13]

    Personalized broad learning system for facial expression.Multimedia Tools and Applications, 79(23):16627–16644, 2020

    Jing Han, Lun Xie, Jing Liu, and Xue Li. Personalized broad learning system for facial expression.Multimedia Tools and Applications, 79(23):16627–16644, 2020. 1

    2020

  6. [14]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. InProceed- ings of the IEEE conference on computer vision and pattern recognition, pages 770–778, Piscataway, USA, 2016. IEEE. 6, 9

    2016

  7. [15]

    Cultural confu- sions show that facial expressions are not universal.Current biology, 19(18):1543–1548, 2009

    Rachael E Jack, Caroline Blais, Christoph Scheepers, Philippe G Schyns, and Roberto Caldara. Cultural confu- sions show that facial expressions are not universal.Current biology, 19(18):1543–1548, 2009. 1

    2009

  8. [16]

    Disentangling identity and pose for facial expression recognition.IEEE Transactions on Affective Computing, 13(4):1868–1878, 2022

    Jing Jiang and Weihong Deng. Disentangling identity and pose for facial expression recognition.IEEE Transactions on Affective Computing, 13(4):1868–1878, 2022. 3

    2022

  9. [17]

    Cfda-csf: A multi-modal domain adaptation method for cross-subject emotion recognition.IEEE Transactions on Af- fective Computing, 15(3):1502–1513, 2024

    Magdiel Jim ´enez-Guarneros and Gibran Fuentes-Pineda. Cfda-csf: A multi-modal domain adaptation method for cross-subject emotion recognition.IEEE Transactions on Af- fective Computing, 15(3):1502–1513, 2024. 2, 3

    2024

  10. [18]

    Magdiel Jim ´enez-Guarneros, Gibran Fuentes-Pineda, and Jonas Grande-Barreto. Mmda: A multimodal and mul- tisource domain adaptation method for cross-subject emo- tion recognition from eeg and eye movement signals.IEEE Transactions on Computational Social Systems, 2024. 2

    2024

  11. [19]

    Contrastive adaptation network for single-and multi-source domain adaptation.IEEE TPAMI, 44(4):1793–1804, 2020

    Guoliang Kang, Lu Jiang, Yunchao Wei, Yi Yang, and Alexander Hauptmann. Contrastive adaptation network for single-and multi-source domain adaptation.IEEE TPAMI, 44(4):1793–1804, 2020. 2, 3, 5, 6, 7, 12, 14, 15

    2020

  12. [20]

    Panns: Large-scale pretrained audio neural networks for audio pattern recogni- tion.IEEE/ACM Transactions on Audio, Speech, and Lan- guage Processing, 28:2880–2894, 2020

    Qiuqiang Kong, Yin Cao, Turab Iqbal, Yuxuan Wang, Wenwu Wang, and Mark D Plumbley. Panns: Large-scale pretrained audio neural networks for audio pattern recogni- tion.IEEE/ACM Transactions on Audio, Speech, and Lan- guage Processing, 28:2880–2894, 2020. 6, 9

    2020

  13. [21]

    Deep emotion transfer network for cross-database facial expression recognition

    Shan Li and Weihong Deng. Deep emotion transfer network for cross-database facial expression recognition. In24th International Conference on Pattern Recognition (ICPR), pages 3092–3099, Piscataway, USA, 2018. IEEE. 1

    2018

  14. [22]

    Wei Liu, Jie-Lin Qiu, Wei-Long Zheng, and Bao-Liang Lu. Settings Methods Target Subjects Sub-1 Sub-2 Sub-3 Sub-4 Sub-5 Sub-6 Sub-7 Sub-8 Sub-9 Sub-10 Avg F1 Lower Bound Visual-only Physio-only Fusion 55.2 36.6 47.0 17.3 45.7 13.0 32.5 37.8 36.4 67.9 22.4 77.4 46.1 37.5 53.4 46...

    2021

  15. [23]

    Unsupervised domain adaptation with residual trans- fer networks.Advances in neural information processing systems, 29, 2016

    Mingsheng Long, Han Zhu, Jianmin Wang, and Michael I Jordan. Unsupervised domain adaptation with residual trans- fer networks.Advances in neural information processing systems, 29, 2016. 1

    2016

  16. [24]

    Conditional adversarial domain adapta- tion.NeurIPS, 31, 2018

    Mingsheng Long, Zhangjie Cao, Jianmin Wang, and Michael I Jordan. Conditional adversarial domain adapta- tion.NeurIPS, 31, 2018. 6, 7, 15

    2018

  17. [25]

    Stem: An approach to multi-source domain adaptation with guarantees

    Van-Anh Nguyen, Tuan Nguyen, Trung Le, Quan Hung Tran, and Dinh Phung. Stem: An approach to multi-source domain adaptation with guarantees. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 9352–9363, Piscataway, USA, 2021. IEEE. 3

    2021

  18. [26]

    Moment matching for multi-source domain adaptation

    Xingchao Peng, Qinxun Bai, Xide Xia, Zijun Huang, Kate Saenko, and Bo Wang. Moment matching for multi-source domain adaptation. InProceedings of the IEEE/CVF inter- national conference on computer vision, pages 1406–1415, Piscataway, USA, 2019. IEEE. 3

    2019

  19. [27]

    A differential entropy estimator for training neural networks

    Georg Pichler, Pierre Jean A Colombo, Malik Boudiaf, G¨unther Koliander, and Pablo Piantanida. A differential entropy estimator for training neural networks. InInter- national Conference on Machine Learning, pages 17691– 17715. PMLR, PMLR, 2022. 3, 4, 9

    2022

  20. [28]

    Deep weakly supervised domain adaptation for pain localization in videos

    R Gnana Praveen, Eric Granger, and Patrick Cardinal. Deep weakly supervised domain adaptation for pain localization in videos. In2020 15th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2020), pages 473–480. IEEE, 2020. 1

    2020

  21. [29]

    A joint cross-attention model for audio-visual fusion in dimensional emotion recognition

    R Gnana Praveen, Wheidima Carneiro de Melo, Nasib Ul- lah, Haseeb Aslam, Osama Zeeshan, Th ´eo Denorme, Marco Pedersoli, Alessandro L Koerich, Simon Bacon, Patrick Car- dinal, et al. A joint cross-attention model for audio-visual fusion in dimensional emotion recognition. InPr...

    2022

  22. [30]

    Deep domain adaptation with ordinal regression for pain assessment using weakly-labeled videos.Image and Vision Computing, 110:104167, 2021

    Gnana Praveen Rajasekhar, Eric Granger, and Patrick Cardi- nal. Deep domain adaptation with ordinal regression for pain assessment using weakly-labeled videos.Image and Vision Computing, 110:104167, 2021. 1

    2021

  23. [31]

    Multi-source unsupervised domain adaptation via pseudo target domain.IEEE Transactions on Image Pro- cessing, 31:2122–2135, 2022

    Chuan-Xian Ren, Yong-Hui Liu, Xi-Wen Zhang, and Ke- Kun Huang. Multi-source unsupervised domain adaptation via pseudo target domain.IEEE Transactions on Image Pro- cessing, 31:2122–2135, 2022. 5, 6

    2022

  24. [32]

    An overview of gradient descent optimiza- tion algorithms.arXiv preprint arXiv:1609.04747, 2016

    Sebastian Ruder. An overview of gradient descent optimiza- tion algorithms.arXiv preprint arXiv:1609.04747, 2016. 9

    2016 arXiv

  25. [33]

    Multi-source domain adaptation via supervised con- trastive learning and confident consistency regularization

    Marin Scalbert, Maria Vakalopoulou, and Florent Couzini ´e- Devy. Multi-source domain adaptation via supervised con- trastive learning and confident consistency regularization. arXiv preprint arXiv:2106.16093, 2021. 2, 3, 5, 6, 7, 13, 14, 15

    2021 arXiv

  26. [34]

    Equivalence of distance-based and rkhs-based statistics in hypothesis testing.The annals of statistics, pages 2263–2291, 2013

    Dino Sejdinovic, Bharath Sriperumbudur, Arthur Gretton, and Kenji Fukumizu. Equivalence of distance-based and rkhs-based statistics in hypothesis testing.The annals of statistics, pages 2263–2291, 2013. 5, 6, 7, 8, 15, 16, 17, 18

    2013

  27. [35]

    Disentan- gled source-free personalization for facial expression recog- nition with neutral target data

    Masoumeh Sharafi, Emma Ollivier, Muhammad Osama Zeeshan, Soufiane Belharbi, Alessandro Lameiras Koerich, Marco Pedersoli, Simon Bacon, and Eric Granger. Disentan- gled source-free personalization for facial expression recog- nition with neutral target data. In2025 IEEE 19th In...

    2025

  28. [36]

    End-to-end mul- timodal emotion recognition using deep neural networks

    Panagiotis Tzirakis, George Trigeorgis, Mihalis A Nicolaou, Bj¨orn W Schuller, and Stefanos Zafeiriou. End-to-end mul- timodal emotion recognition using deep neural networks. Journal of selected topics in signal processing, 11(8):1301– 1309, 2017. 1, 2

    2017

  29. [37]

    Visualizing data using t-sne.Journal of machine learning research, 9 (11), 2008

    Laurens Van der Maaten and Geoffrey Hinton. Visualizing data using t-sne.Journal of machine learning research, 9 (11), 2008. 8

    2008

  30. [38]

    Your classifier can secretly suffice multi-source domain adaptation.Advances in Neural Infor- mation Processing Systems, 33:4647–4659, 2020

    Naveen Venkat, Jogendra Nath Kundu, Durgesh Singh, Am- bareesh Revanur, et al. Your classifier can secretly suffice multi-source domain adaptation.Advances in Neural Infor- mation Processing Systems, 33:4647–4659, 2020. 3

    2020

  31. [39]

    Joint multimodal transformer for emotion recognition in the wild

    Paul Waligora, Muhammad Haseeb Aslam, Muham- mad Osama Zeeshan, Soufiane Belharbi, Alessan- dro Lameiras Koerich, Marco Pedersoli, Simon Bacon, and Eric Granger. Joint multimodal transformer for emotion recognition in the wild. InProceedings of the IEEE/CVF Conference on Compu...

    2024

  32. [40]

    The biovid heat pain database data for the advance- ment and systematic validation of an automated pain recog- nition system

    Steffen Walter, Sascha Gruss, Hagen Ehleiter, Junwen Tan, Harald C Traue, Philipp Werner, Ayoub Al-Hamadi, Stephen Crawcour, Adriano O Andrade, and Gustavo Moreira da Silva. The biovid heat pain database data for the advance- ment and systematic validation of an automated pain...

    2013

  33. [41]

    Multi-modal domain adaptation variational autoencoder for eeg-based emotion recognition

    Yixin Wang, Shuang Qiu, Dan Li, Changde Du, Bao-Liang Lu, and Huiguang He. Multi-modal domain adaptation variational autoencoder for eeg-based emotion recognition. IEEE/CAA Journal of Automatica Sinica, 9(9):1612–1626,

  34. [42]

    Analysis of facial expressiveness during experimentally in- duced heat pain

    Philipp Werner, Ayoub Al-Hamadi, and Steffen Walter. Analysis of facial expressiveness during experimentally in- duced heat pain. In2017 Seventh international conference on affective computing and intelligent interaction workshops and demos (ACIIW), pages 176–180. IEEE, 2017. 6, 10

    2017

  35. [43]

    Uncertainty-aware multi-view co-training for semi-supervised medical image segmentation and do- main adaptation.Medical image analysis, 65:101766, 2020

    Yingda Xia, Dong Yang, Zhiding Yu, Fengze Liu, Jinzheng Cai, Lequan Yu, Zhuotun Zhu, Daguang Xu, Alan Yuille, and Holger Roth. Uncertainty-aware multi-view co-training for semi-supervised medical image segmentation and do- main adaptation.Medical image analysis, 65:101766, 2020. 3

    2020

  36. [44]

    Facial expression recognition with two-branch disentangled generative adver- sarial network.IEEE Transactions on Circuits and Systems for Video Technology, 31(6):2359–2371, 2020

    Siyue Xie, Haifeng Hu, and Yizhen Chen. Facial expression recognition with two-branch disentangled generative adver- sarial network.IEEE Transactions on Circuits and Systems for Video Technology, 31(6):2359–2371, 2020. 3

    2020

  37. [45]

    Deep cocktail network: Multi-source unsuper- vised domain adaptation with category shift

    Ruijia Xu, Ziliang Chen, Wangmeng Zuo, Junjie Yan, and Liang Lin. Deep cocktail network: Multi-source unsuper- vised domain adaptation with category shift. InProceed- ings of the IEEE conference on computer vision and pattern recognition, pages 3964–3973, 2018. 1, 5

    2018

  38. [46]

    Deep co-training with task decomposition for semi- supervised domain adaptation

    Luyu Yang, Yan Wang, Mingfei Gao, Abhinav Shrivas- tava, Kilian Q Weinberger, Wei-Lun Chao, and Ser-Nam Lim. Deep co-training with task decomposition for semi- supervised domain adaptation. InProceedings of the IEEE/CVF international conference on computer vision, pages 8906–8...

    2021

  39. [47]

    Subject-based domain adaptation for facial expression recognition

    Muhammad Osama Zeeshan, Muhammad Haseeb Aslam, Soufiane Belharbi, Alessandro Lameiras Koerich, Marco Pedersoli, Simon Bacon, and Eric Granger. Subject-based domain adaptation for facial expression recognition. In2024 IEEE 18th International Conference on Automatic Face and Ges...

    2024

  40. [48]

    Progressive multi- source domain adaptation for personalized facial expression recognition.IEEE Transactions on Affective Computing,

    Muhammad Osama Zeeshan, Marco Pedersoli, Alessandro Lameiras Koerich, and Eric Granger. Progressive multi- source domain adaptation for personalized facial expression recognition.IEEE Transactions on Affective Computing,

  41. [49]

    Emotion recognition using multi-modal data and machine learning techniques: A tutorial and review.Information Fu- sion, 59:103–126, 2020

    Jianhua Zhang, Zhong Yin, Peng Chen, and Stefano Nichele. Emotion recognition using multi-modal data and machine learning techniques: A tutorial and review.Information Fu- sion, 59:103–126, 2020. 1

    2020

  42. [50]

    Madan: multi-source adversarial domain aggregation network for domain adaptation.IJCV, 129(8):2399–2424, 2021

    Sicheng Zhao, Bo Li, Pengfei Xu, Xiangyu Yue, Guiguang Ding, and Kurt Keutzer. Madan: multi-source adversarial domain aggregation network for domain adaptation.IJCV, 129(8):2399–2424, 2021. 3

    2021

  43. [51]

    Discrimina- tive feature adaptation for cross-domain facial expression recognition

    Ronghang Zhu, Gaoli Sang, and Qijun Zhao. Discrimina- tive feature adaptation for cross-domain facial expression recognition. In2016 International Conference on Biomet- rics (ICB), pages 1–7. IEEE, 2016. 1

    2016

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.