pith. sign in

arxiv: 2606.23758 · v2 · pith:7SPK7UVZnew · submitted 2026-06-22 · 💻 cs.LG · cs.AI

Exploring Dualistic Meta-Learning to Enhance Domain Generalization in Open Set Scenarios

Xiran Wang , Jian Zhang , Lei Qi , Yang Gao , Yinghuan Shi This is my paper

Pith reviewed 2026-06-26 09:16 UTC · model grok-4.3

classification 💻 cs.LG cs.AI
keywords domain generalizationopen set recognitionmeta-learningone-vs-all classifiersdomain shiftout-of-distribution detectionclass imbalance
0
0 comments X

The pith

MEDIC uses simultaneous inter-domain and inter-class gradient matching in meta-learning to balance one-vs-all classifier boundaries for open set domain generalization.

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

Domain generalization struggles when target domains introduce unseen classes, because one-vs-all classifiers trained on source data suffer from severe positive-negative sample imbalance that skews boundaries toward the positives. This skew causes over-rejection of known classes from new domains. The paper introduces MEDIC, a meta-learning strategy that performs implicit gradient matching on both inter-domain and inter-class task splits at once. The dual matching is intended to locate decision boundaries that remain balanced across both kinds of variation. Experiments show improved open-set accuracy while preserving competitive closed-set generalization.

Core claim

By considering implicit gradient matching towards inter-domain and inter-class task splits simultaneously, MEDIC finds optimal boundaries balanced for both domains and classes, correcting the skew that one-vs-all classifiers exhibit under sample imbalance in open set domain generalization.

What carries the argument

dualistic meta-learning with joint domain-class matching (MEDIC), which performs simultaneous implicit gradient matching on inter-domain and inter-class splits to produce balanced boundaries.

If this is right

  • MEDIC outperforms prior open-set domain generalization methods.
  • The approach maintains competitive performance on standard closed-set domain generalization tasks.
  • It reduces over-rejection of out-of-distribution samples that belong to known classes.
  • The same meta-learning procedure can be applied when both domain and label shifts are present.

Where Pith is reading between the lines

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

  • The dual-matching idea might transfer to other multi-objective meta-learning problems that involve simultaneous shifts of different types.
  • If the balancing effect holds, similar joint-split training could be tested in federated or continual learning settings with both domain and class imbalance.
  • The method suggests exploring whether explicit rather than implicit gradient matching yields further gains when the number of splits increases.

Load-bearing premise

That sample imbalance is the main cause of skewed boundaries in one-vs-all classifiers and that simultaneous gradient matching on domain and class splits will produce balanced boundaries without creating new problems.

What would settle it

An ablation experiment in which the joint domain-class matching component is removed yet open-set performance remains unchanged, or a controlled setting where balanced boundaries appear without the dualistic split.

Figures

Figures reproduced from arXiv: 2606.23758 by Jian Zhang, Lei Qi, Xiran Wang, Yang Gao, Yinghuan Shi.

Figure 1
Figure 1. Figure 1: An example for the variation of decision boundaries of a [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Previous research [16] has demonstrated that the large angle between [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Overview of MEDIC during one training iteration. [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Comparison of different learning strategies. (a) A single step. (b) (c) Multiple tasks per step. (d) Maximum steps. A greater number of steps implies [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: (a) (b) (c) Standardized mean differences of class feature pairs. Warmer colors indicate more distinct features. It can be observed that the one task per [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: An example of adaptive task sampling. Initially, class 2 is randomly [PITH_FULL_IMAGE:figures/full_fig_p007_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Supplementary visual aids for positive correlation between task-wise [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗
Figure 9
Figure 9. Figure 9: Training time (sec) and memory cost (mb) with respect to the number of steps during 5000 iterations. TABLE XIII CONFIDENCE SCORES (%) ON PACS / SKETCH. Method ERM MLDG Fish MEDIC MEDIC++ confp 75.59 77.30 75.76 82.23 85.08 confn 84.29 84.04 85.12 82.12 85.15 Therefore, |λ − 0.5| is minimized when Eq. (36) holds. This suggests that balanced output across known classes serves as a sufficient condition for in… view at source ↗
Figure 10
Figure 10. Figure 10: T-SNE results of feature representations in the target domain, where pink and green corresponds to known and unknown classes respectively. [PITH_FULL_IMAGE:figures/full_fig_p014_10.png] view at source ↗
read the original abstract

Domain generalization learns from multiple source domains to generalize to unseen target domains. However, it often neglects the realistic case of label mismatch between source and target. Open set domain generalization is then proposed to recognize unseen classes in unseen domains. A simple approach trains one-vs-all classifiers to separate each class and detect outliers as unknown. Yet, the imbalance between few positive samples and many negative samples skews the decision boundary towards the positive ones, leading the model to over-reject out-of-distribution data, even from known classes in unseen domains. In this paper, we propose a novel meta-learning stategy called dualistic MEta-learning with joint DomaIn-Class matching (MEDIC), which considers implicit gradient matching towards inter-domain and inter-class task splits simultaneously to find optimal boundaries balanced for both domains and classes. Experimental results show that MEDIC not only outperforms prior methods in open set scenarios, but also maintains competitive close set generalization ability.

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.

Referee Report

0 major / 2 minor

Summary. The manuscript proposes MEDIC, a dualistic meta-learning strategy for open set domain generalization. It identifies that one-vs-all classifiers suffer from positive-negative sample imbalance that skews decision boundaries toward positives, causing over-rejection of known classes from unseen domains. MEDIC performs implicit gradient matching simultaneously on inter-domain and inter-class task splits to produce boundaries balanced for both, and reports experimental outperformance over prior methods in open-set settings while remaining competitive on closed-set generalization.

Significance. If the central claim holds, the work supplies a mechanistic meta-learning solution that jointly targets domain and class imbalance via gradient matching, which could strengthen robustness in realistic open-set DG scenarios. The explicit construction linking the identified skew problem to dualistic matching is a clear strength of the approach as described.

minor comments (2)
  1. Ensure that the full methods section provides the precise formulation of the joint implicit gradient matching objective (including any weighting between domain and class splits) so that the balancing mechanism can be directly inspected and reproduced.
  2. The abstract states that MEDIC 'maintains competitive close set generalization ability'; include quantitative tables comparing closed-set accuracy against the same baselines used for open-set evaluation to support this claim.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the summary of our MEDIC approach for open-set domain generalization. The report provides no specific major comments despite the 'uncertain' recommendation, so we have no points to address point-by-point at this time. We remain available to respond to any additional feedback or concerns the referee may have.

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper presents MEDIC as a novel meta-learning strategy that applies implicit gradient matching simultaneously to inter-domain and inter-class task splits to address boundary skew in one-vs-all classifiers for open-set domain generalization. The abstract and description supply a direct mechanistic motivation without any equations, fitted parameters renamed as predictions, or self-citation chains that reduce the central claim to its own inputs by construction. The method is constructed to target the stated imbalance problem rather than deriving from prior results of the same authors or re-labeling known patterns. This is the most common honest finding for a method paper whose core contribution is an explicit algorithmic design choice.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

With only the abstract available, no specific free parameters, axioms, or invented entities can be identified from the text.

pith-pipeline@v0.9.1-grok · 5695 in / 932 out tokens · 24382 ms · 2026-06-26T09:16:55.289187+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

101 extracted references · 8 canonical work pages · 5 internal anchors

  1. [1]

    Research on overfitting of deep learning,

    H. Li, J. Li, X. Guan, B. Liang, Y . Lai, and X. Luo, “Research on overfitting of deep learning,” in2019 15th International Conference on Computational Intelligence and Security, 2019, pp. 78–81

    2019

  2. [2]

    Generalizing to unseen domains: A survey on domain generalization,

    J. Wang, C. Lan, C. Liu, Y . Ouyang, T. Qin, W. Lu, Y . Chen, W. Zeng, and P. Yu, “Generalizing to unseen domains: A survey on domain generalization,”IEEE Transactions on Knowledge and Data Engineering, 2022

    2022

  3. [3]

    Learning to generalize: Meta-learning for domain generalization,

    D. Li, Y . Yang, Y .-Z. Song, and T. Hospedales, “Learning to generalize: Meta-learning for domain generalization,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 32, no. 1, 2018

    2018

  4. [4]

    Episodic training for domain generalization,

    D. Li, J. Zhang, Y . Yang, C. Liu, Y .-Z. Song, and T. M. Hospedales, “Episodic training for domain generalization,” inProceedings of the IEEE/CVF International Conference on Computer Vision, 2019, pp. 1446–1455

    2019

  5. [5]

    Domain generalization with mixstyle,

    K. Zhou, Y . Yang, Y . Qiao, and T. Xiang, “Domain generalization with mixstyle,” inInternational Conference on Learning Representations, 2020

    2020

  6. [6]

    Understanding hessian alignment for domain generalization,

    S. Hemati, G. Zhang, A. Estiri, and X. Chen, “Understanding hessian alignment for domain generalization,” inProceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 19 004– 19 014

    2023

  7. [7]

    Toward open set recognition,

    W. J. Scheirer, A. de Rezende Rocha, A. Sapkota, and T. E. Boult, “Toward open set recognition,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 35, no. 7, pp. 1757–1772, 2012

    2012

  8. [8]

    Clinical research for rare disease: opportunities, challenges, and solutions,

    R. C. Griggs, M. Batshaw, M. Dunkle, R. Gopal-Srivastava, E. Kaye, J. Krischer, T. Nguyen, K. Paulus, P. A. Merkelet al., “Clinical research for rare disease: opportunities, challenges, and solutions,”Molecular genetics and metabolism, vol. 96, no. 1, pp. 20–26, 2009

    2009

  9. [9]

    Open domain gener- alization with domain-augmented meta-learning,

    Y . Shu, Z. Cao, C. Wang, J. Wang, and M. Long, “Open domain gener- alization with domain-augmented meta-learning,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 9624–9633

    2021

  10. [10]

    Open-set domain generalization via metric learning,

    K. Katsumata, I. Kishida, A. Amma, and H. Nakayama, “Open-set domain generalization via metric learning,” in2021 IEEE International Conference on Image Processing, 2021, pp. 459–463

    2021

  11. [11]

    Con- ditional variational capsule network for open set recognition,

    Y . Guo, G. Camporese, W. Yang, A. Sperduti, and L. Ballan, “Con- ditional variational capsule network for open set recognition,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 103–111

    2021

  12. [12]

    Conditional gaussian distribution learning for open set recognition,

    X. Sun, Z. Yang, C. Zhang, K.-V . Ling, and G. Peng, “Conditional gaussian distribution learning for open set recognition,” inProceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 13 480–13 489

    2020

  13. [13]

    Ovanet: One-vs-all network for universal domain adaptation,

    K. Saito and K. Saenko, “Ovanet: One-vs-all network for universal domain adaptation,” inProceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 9000–9009

    2021

  14. [14]

    Separate to adapt: Open set domain adaptation via progressive separation,

    H. Liu, Z. Cao, M. Long, J. Wang, and Q. Yang, “Separate to adapt: Open set domain adaptation via progressive separation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 2927–2936

    2019

  15. [15]

    Meta- learning in neural networks: A survey,

    T. Hospedales, A. Antoniou, P. Micaelli, and A. Storkey, “Meta- learning in neural networks: A survey,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 9, pp. 5149–5169, 2021

    2021

  16. [16]

    Gradient matching for domain generalization,

    Y . Shi, J. Seely, P. Torr, N. Siddharth, A. Hannun, N. Usunier, and G. Synnaeve, “Gradient matching for domain generalization,” in International Conference on Learning Representations, 2021

    2021

  17. [17]

    Generalizable decision bound- aries: Dualistic meta-learning for open set domain generalization,

    X. Wang, J. Zhang, L. Qi, and Y . Shi, “Generalizable decision bound- aries: Dualistic meta-learning for open set domain generalization,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 11 564–11 573

    2023

  18. [18]

    On First-Order Meta-Learning Algorithms

    A. Nichol, J. Achiam, and J. Schulman, “On first-order meta-learning algorithms,”arXiv preprint arXiv:1803.02999, 2018

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  19. [19]

    Domain generalization with adversarial feature learning,

    H. Li, S. J. Pan, S. Wang, and A. C. Kot, “Domain generalization with adversarial feature learning,” inProceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 5400–5409

    2018

  20. [20]

    Domain-adversarial training of neural networks,

    Y . Ganin, E. Ustinova, H. Ajakan, P. Germain, H. Larochelle, F. Lavio- lette, M. Marchand, and V . Lempitsky, “Domain-adversarial training of neural networks,”The Journal of Machine Learning Research, vol. 17, no. 1, pp. 2096–2030, 2016

    2096

  21. [21]

    Domain adversarial neural networks for domain generalization: When it works and how to improve,

    A. Sicilia, X. Zhao, and S. J. Hwang, “Domain adversarial neural networks for domain generalization: When it works and how to improve,”Machine Learning, pp. 1–37, 2023

    2023

  22. [22]

    Domain adversarial active learning for domain generalization classification,

    J. Chen, L. Ding, Y . Yang, Z. Di, and Y . Xiang, “Domain adversarial active learning for domain generalization classification,”IEEE Transac- tions on Knowledge and Data Engineering, vol. 37, no. 1, pp. 226–238, 2024

    2024

  23. [23]

    Invariant risk minimization,

    M. Arjovsky, L. Bottou, I. Gulrajani, and D. Lopez-Paz, “Invariant risk minimization,”Stat, vol. 1050, p. 27, 2020

    2020

  24. [24]

    Invariance principle meets information bottleneck for out-of-distribution generalization,

    K. Ahuja, E. Caballero, D. Zhang, J.-C. Gagnon-Audet, Y . Bengio, I. Mitliagkas, and I. Rish, “Invariance principle meets information bottleneck for out-of-distribution generalization,”Advances in Neural Information Processing Systems, vol. 34, pp. 3438–3450, 2021

    2021

  25. [25]

    Learning to balance specificity and invariance for in and out of domain generalization,

    P. Chattopadhyay, Y . Balaji, and J. Hoffman, “Learning to balance specificity and invariance for in and out of domain generalization,” in European Conference on Computer Vision, 2020, pp. 301–318

    2020

  26. [26]

    Causality inspired representation learning for domain generalization,

    F. Lv, J. Liang, S. Li, B. Zang, C. H. Liu, Z. Wang, and D. Liu, “Causality inspired representation learning for domain generalization,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 8046–8056

    2022

  27. [27]

    Heterogeneous domain generalization via domain mixup,

    Y . Wang, H. Li, and A. C. Kot, “Heterogeneous domain generalization via domain mixup,” inIEEE International Conference on Acoustics, Speech and Signal Processing, 2020, pp. 3622–3626

    2020

  28. [28]

    A fourier- based framework for domain generalization,

    Q. Xu, R. Zhang, Y . Zhang, Y . Wang, and Q. Tian, “A fourier- based framework for domain generalization,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 14 383–14 392

    2021

  29. [29]

    Aloft: A lightweight mlp-like architecture with dynamic low-frequency transform for domain gener- 15 alization,

    J. Guo, N. Wang, L. Qi, and Y . Shi, “Aloft: A lightweight mlp-like architecture with dynamic low-frequency transform for domain gener- 15 alization,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 24 132–24 141

    2023

  30. [30]

    Progressive domain expansion network for single domain gen- eralization,

    L. Li, K. Gao, J. Cao, Z. Huang, Y . Weng, X. Mi, Z. Yu, X. Li, and B. Xia, “Progressive domain expansion network for single domain gen- eralization,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 224–233

    2021

  31. [31]

    Learning to generate novel domains for domain generalization,

    K. Zhou, Y . Yang, T. Hospedales, and T. Xiang, “Learning to generate novel domains for domain generalization,” inEuropean Conference on Computer Vision, 2020, pp. 561–578

    2020

  32. [32]

    A simple feature augmentation for domain generalization,

    P. Li, D. Li, W. Li, S. Gong, Y . Fu, and T. M. Hospedales, “A simple feature augmentation for domain generalization,” inProceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 8886–8895

    2021

  33. [33]

    Feature-based style randomization for domain generalization,

    Y . Wang, L. Qi, Y . Shi, and Y . Gao, “Feature-based style randomization for domain generalization,”IEEE Transactions on Circuits and Systems for Video Technology, vol. 32, no. 8, pp. 5495–5509, 2022

    2022

  34. [34]

    Mvdg: A unified multi-view framework for domain generalization,

    J. Zhang, L. Qi, Y . Shi, and Y . Gao, “Mvdg: A unified multi-view framework for domain generalization,” inEuropean Conference on Computer Vision, 2022, pp. 161–177

    2022

  35. [35]

    Do- main generalization via model-agnostic learning of semantic features,

    Q. Dou, D. Coelho de Castro, K. Kamnitsas, and B. Glocker, “Do- main generalization via model-agnostic learning of semantic features,” Advances in Neural Information Processing Systems, vol. 32, 2019

    2019

  36. [36]

    Metareg: Towards domain generalization using meta-regularization,

    Y . Balaji, S. Sankaranarayanan, and R. Chellappa, “Metareg: Towards domain generalization using meta-regularization,”Advances in Neural Information Processing Systems, vol. 31, 2018

    2018

  37. [37]

    Domain adaptive ensemble learning,

    K. Zhou, Y . Yang, Y . Qiao, and T. Xiang, “Domain adaptive ensemble learning,”IEEE Transactions on Image Processing, vol. 30, pp. 8008– 8018, 2021

    2021

  38. [38]

    Swad: Domain generalization by seeking flat minima,

    J. Cha, S. Chun, K. Lee, H.-C. Cho, S. Park, Y . Lee, and S. Park, “Swad: Domain generalization by seeking flat minima,”Advances in Neural Information Processing Systems, vol. 34, pp. 22 405–22 418, 2021

    2021

  39. [39]

    Ensemble of averages: Improving model selection and boosting performance in domain gener- alization,

    D. Arpit, H. Wang, Y . Zhou, and C. Xiong, “Ensemble of averages: Improving model selection and boosting performance in domain gener- alization,”Advances in Neural Information Processing Systems, vol. 35, pp. 8265–8277, 2022

    2022

  40. [40]

    Privacy-preserving con- strained domain generalization via gradient alignment,

    C. X. Tian, H. Li, Y . Wang, and S. Wang, “Privacy-preserving con- strained domain generalization via gradient alignment,”IEEE Transac- tions on Knowledge and Data Engineering, vol. 36, no. 5, pp. 2142– 2150, 2023

    2023

  41. [41]

    Self-challenging improves cross-domain generalization,

    Z. Huang, H. Wang, E. P. Xing, and D. Huang, “Self-challenging improves cross-domain generalization,” inEuropean Conference on Computer Vision, 2020, pp. 124–140

    2020

  42. [42]

    Domain generalization model of deep convolutional networks based on sand-mask,

    J. Wang, L. Chen, and R. Wang, “Domain generalization model of deep convolutional networks based on sand-mask,”Algorithms, vol. 15, no. 6, p. 215, 2022

    2022

  43. [43]

    Domain generalization via gradient surgery,

    L. Mansilla, R. Echeveste, D. H. Milone, and E. Ferrante, “Domain generalization via gradient surgery,” inProceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 6630–6638

    2021

  44. [44]

    Deep visual domain adaptation: A survey,

    M. Wang and W. Deng, “Deep visual domain adaptation: A survey,” Neurocomputing, vol. 312, pp. 135–153, 2018

    2018

  45. [45]

    Recent advances in open set recognition: A survey,

    C. Geng, S.-j. Huang, and S. Chen, “Recent advances in open set recognition: A survey,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 10, pp. 3614–3631, 2020

    2020

  46. [46]

    Reducing network agnosto- phobia,

    A. R. Dhamija, M. G ¨unther, and T. Boult, “Reducing network agnosto- phobia,”Advances in Neural Information Processing Systems, vol. 31, 2018

    2018

  47. [47]

    Deep anomaly detec- tion with outlier exposure,

    D. Hendrycks, M. Mazeika, and T. Dietterich, “Deep anomaly detec- tion with outlier exposure,” inInternational Conference on Learning Representations, 2018

    2018

  48. [48]

    Generative openmax for multi-class open set classification,

    Z. Ge, S. Demyanov, Z. Chen, and R. Garnavi, “Generative openmax for multi-class open set classification,” inBritish Machine Vision Conference 2017, 2017

    2017

  49. [49]

    Open set learning with counterfactual images,

    L. Neal, M. Olson, X. Fern, W.-K. Wong, and F. Li, “Open set learning with counterfactual images,” inProceedings of the European Conference on Computer Vision, 2018, pp. 613–628

    2018

  50. [50]

    Opengan: Open-set recognition via open data generation,

    S. Kong and D. Ramanan, “Opengan: Open-set recognition via open data generation,” inProceedings of the IEEE/CVF International Con- ference on Computer Vision, 2021, pp. 813–822

    2021

  51. [51]

    Towards open set deep networks,

    A. Bendale and T. E. Boult, “Towards open set deep networks,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016, pp. 1563–1572

    2016

  52. [52]

    Extreme value theory,

    R. L. Smith, “Extreme value theory,”Handbook of Applicable Mathe- matics, vol. 7, pp. 437–471, 1990

    1990

  53. [53]

    C2ae: Class conditioned auto-encoder for open-set recognition,

    P. Oza and V . M. Patel, “C2ae: Class conditioned auto-encoder for open-set recognition,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 2307–2316

    2019

  54. [54]

    Classification-reconstruction learning for open-set recognition,

    R. Yoshihashi, W. Shao, R. Kawakami, S. You, M. Iida, and T. Nae- mura, “Classification-reconstruction learning for open-set recognition,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 4016–4025

    2019

  55. [55]

    Spectral-spatial latent reconstruction for open-set hyperspectral image classification,

    J. Yue, L. Fang, and M. He, “Spectral-spatial latent reconstruction for open-set hyperspectral image classification,”IEEE Transactions on Image Processing, vol. 31, pp. 5227–5241, 2022

    2022

  56. [56]

    Adversarial reciprocal points learning for open set recognition,

    G. Chen, P. Peng, X. Wang, and Y . Tian, “Adversarial reciprocal points learning for open set recognition,”IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021

    2021

  57. [57]

    Pmal: Open set recognition via robust prototype mining,

    J. Lu, Y . Xu, H. Li, Z. Cheng, and Y . Niu, “Pmal: Open set recognition via robust prototype mining,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 36, no. 2, 2022, pp. 1872–1880

    2022

  58. [58]

    Thrun and L

    S. Thrun and L. Pratt,Learning to learn. Springer Science & Business Media, 2012

    2012

  59. [59]

    On data efficiency of meta-learning,

    M. Al-Shedivat, L. Li, E. Xing, and A. Talwalkar, “On data efficiency of meta-learning,” inInternational Conference on Artificial Intelligence and Statistics. PMLR, 2021, pp. 1369–1377

    2021

  60. [60]

    Model-agnostic meta-learning for fast adaptation of deep networks,

    C. Finn, P. Abbeel, and S. Levine, “Model-agnostic meta-learning for fast adaptation of deep networks,” inInternational Conference on Machine Learning. PMLR, 2017, pp. 1126–1135

    2017

  61. [61]

    One ring to bring them all: Towards open-set recognition under domain shift,

    S. Yang, Y . Wang, K. Wang, S. Jui, and J. van de Weijer, “One ring to bring them all: Towards open-set recognition under domain shift,” ArXiv Preprint ArXiv:2206.03600, 2022

    work page arXiv 2022

  62. [62]

    Activate and reject: Towards safe domain generalization under cate- gory shift,

    C. Chen, L. Tang, L. Tao, H.-Y . Zhou, Y . Huang, X. Han, and Y . Yu, “Activate and reject: Towards safe domain generalization under cate- gory shift,” inProceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 11 552–11 563

    2023

  63. [63]

    Crossmatch: Cross-classifier consistency regular- ization for open-set single domain generalization,

    R. Zhu and S. Li, “Crossmatch: Cross-classifier consistency regular- ization for open-set single domain generalization,” inInternational Conference on Learning Representations, 2021

    2021

  64. [64]

    Unknown prompt the only lacuna: Unveiling clip’s potential for open domain generalization,

    M. Singha, A. Jha, S. Bose, A. Nair, M. Abdar, and B. Banerjee, “Unknown prompt the only lacuna: Unveiling clip’s potential for open domain generalization,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 13 309–13 319

    2024

  65. [65]

    Meta- learning to teach semantic prompts for open domain generalization in vision-language models,

    S. Bose, M. Singha, A. Jha, S. Mukhopadhyay, and B. Banerjee, “Meta- learning to teach semantic prompts for open domain generalization in vision-language models,”Transactions on Machine Learning Research, 2025

    2025

  66. [66]

    Osloprompt: Bridging low-supervision challenges and open-set domain generalization in clip,

    D. Gupta, M. Singha, S. B. Rongali, A. Jha, M. H. Khan, B. Banerjee et al., “Osloprompt: Bridging low-supervision challenges and open-set domain generalization in clip,” inProceedings of the Computer Vision and Pattern Recognition Conference, 2025, pp. 10 110–10 120

    2025

  67. [67]

    High-resolution image synthesis with latent diffusion models,

    R. Rombach, A. Blattmann, D. Lorenz, P. Esser, and B. Ommer, “High-resolution image synthesis with latent diffusion models,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 10 684–10 695

    2022

  68. [68]

    GPT-4 Technical Report

    J. Achiam, S. Adler, S. Agarwal, L. Ahmad, I. Akkaya, F. L. Aleman, D. Almeida, J. Altenschmidt, S. Altman, S. Anadkatet al., “Gpt-4 technical report,”arXiv preprint arXiv:2303.08774, 2023

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  69. [69]

    Learning transferable visual models from natural language supervision,

    A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clarket al., “Learning transferable visual models from natural language supervision,” inInternational conference on machine learning. PmLR, 2021, pp. 8748–8763

    2021

  70. [70]

    Learning to transport for open set domain generalization,

    C. Li, S. Wang, Y . Long, and H. Zhang, “Learning to transport for open set domain generalization,”Pattern Recognition, p. 112988, 2025

    2025

  71. [71]

    Advancing open-set domain gener- alization using evidential bi-level hardest domain scheduler,

    K. Peng, D. Wen, K. Yang, A. Luo, Y . Chen, J. Fu, M. S. Sarfraz, A. Roitberg, and R. Stiefelhagen, “Advancing open-set domain gener- alization using evidential bi-level hardest domain scheduler,”Advances in Neural Information Processing Systems, vol. 37, pp. 85 412–85 440, 2024

    2024

  72. [72]

    Mitigating label noise using prompt-based hyperbolic meta-learning in open-set domain generalization,

    K. Peng, D. Wen, M. S. Sarfraz, Y . Chen, J. Zheng, D. Schneider, K. Yang, J. Wu, A. Roitberg, and R. Stiefelhagen, “Mitigating label noise using prompt-based hyperbolic meta-learning in open-set domain generalization,”International Journal of Computer Vision, vol. 134, no. 3, p. 99, 2026

    2026

  73. [73]

    An overview of gradient descent optimization algorithms

    S. Ruder, “An overview of gradient descent optimization algorithms,” arXiv preprint arXiv:1609.04747, 2016

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  74. [74]

    On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima

    N. S. Keskar, D. Mudigere, J. Nocedal, M. Smelyanskiy, and P. T. P. Tang, “On large-batch training for deep learning: Generalization gap and sharp minima,”arXiv preprint arXiv:1609.04836, 2016

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  75. [75]

    Neural networks for machine learning lecture 6a overview of mini-batch gradient descent,

    G. Hinton, N. Srivastava, and K. Swersky, “Neural networks for machine learning lecture 6a overview of mini-batch gradient descent,” Cited on, vol. 14, no. 8, p. 2, 2012

    2012

  76. [76]

    Three Factors Influencing Minima in SGD

    S. Jastrzkebski, Z. Kenton, D. Arpit, N. Ballas, A. Fischer, Y . Bengio, and A. Storkey, “Three factors influencing minima in sgd,”arXiv preprint arXiv:1711.04623, 2017. 16

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  77. [77]

    Adaptive task sampling for meta-learning,

    C. Liu, Z. Wang, D. Sahoo, Y . Fang, K. Zhang, and S. C. Hoi, “Adaptive task sampling for meta-learning,” inProceedings of the European Conference on Computer Vision (ECCV), 2020, pp. 752–769

    2020

  78. [78]

    Sequential reptile: inter- task gradient alignment for multilingual learning,

    S. Lee, H. B. Lee, J. Lee, and S. J. Hwang, “Sequential reptile: inter- task gradient alignment for multilingual learning,” inTenth Interna- tional Conference on Learning Representations, 2022

    2022

  79. [79]

    Statistical learning theory,

    V . V . Naumovich and V . Vlamimir, “Statistical learning theory,” 1998

    1998

  80. [80]

    Deeper, broader and artier domain generalization,

    D. Li, Y . Yang, Y .-Z. Song, and T. M. Hospedales, “Deeper, broader and artier domain generalization,” inProceedings of the IEEE International Conference on Computer Vision, 2017, pp. 5542–5550

    2017

Showing first 80 references.