Pith. sign in

REVIEW 4 major objections 5 minor 28 references

PSDA claims that pairing source–target observations—and pairing those pairs—via linear-assignment problems minimizes expected gradient variance for MMD and CORAL, making classical distribution matching competitive with modern UDA at low cos

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-01 10:02 UTC pith:FQXSTUQU

load-bearing objection A genuinely new sampling idea with believable empirical results, but the paper's central variance-minimization theorem is not established as written. the 4 major comments →

arxiv 2607.20367 v1 pith:FQXSTUQU submitted 2026-07-22 cs.LG

Variance-reduced Domain Adaptation using Paired Sampling

Andrea Napoli This is my paper
classification cs.LG
keywords unsupervised domain adaptationvariance reductionmaximum mean discrepancyCORALlinear assignmentminibatch samplinggradient variancepaired sampling
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper sets out to make two classical distribution-matching losses for unsupervised domain adaptation—MMD and CORAL—usable in minibatch training by directly attacking the noise in their gradient estimates. Its proposal, PSDA, builds minibatches from source–target pairs and from pairs of such pairs (quadruplets), where the pairings are chosen to minimise the expected squared error of the discrepancy estimate. Because each pairing step is a linear assignment problem, the authors argue the scheme finds globally optimal pairings cheaply, unlike previous ordering-based variance reduction. Simulations show substantially lower estimator variance at small minibatch sizes, and experiments on three domain-shift benchmarks show that classical MMD/CORAL becomes competitive with, or better than, far more elaborate adversarial UDA methods.

Core claim

The central claim is that the expected squared gradient error of MMD and CORAL, which lack finite-sum structure and are incompatible with classical SVR, can be minimised by inducing dependencies among sampled observations: first match source and target examples so that their co-occurrence minimises the conditional error, then match those pairs into quadruplets that are always sampled together. The paper derives closed-form cost matrices for both stages and shows each stage reduces to a linear assignment problem solvable to global optimality. It reports that this variance reduction is large for small minibatches, transfers to higher target accuracy in realistic training, and costs far less wa

What carries the argument

Two cost matrices: C^st_ij, the expected squared discrepancy error conditional on source example i and target example j being in the minibatch, and C^MM_rq, the same quantity conditional on two matched pairs (r and q) co-occurring. Since sums of such conditional costs recover the unconditional expectation, the matchings are solved as linear assignment problems, producing quadruplets (two source, two target) always sampled together; the closed forms use conditional kernel means for MMD and Welford-style covariance updates for CORAL.

Load-bearing premise

The load-bearing premise is that minimising the sum of per-pair conditional squared-error costs, each computed as ||E[hat D - D | pair]||², minimises the true expected squared gradient error of a full minibatch; this drops the conditional variance term inside each pair and ignores variance from interactions among multiple pairs co-occurring when the minibatch has k>2.

What would settle it

For a tiny synthetic problem (for example 8 source and 8 target points with k=4), enumerate every possible minibatch and every possible pairing, compute the exact expected squared error E||hat D - D||² over all draws, and compare PSDA's pairings with the true global minimiser. If any non-PSDA minibatch distribution has lower variance, the central 'minimise expected gradient variance' claim fails.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • If the claim holds, a drop-in change of sampler—no architecture or loss changes—brings vanilla MMD and CORAL up to par with modern adversarial UDA methods on standard benchmarks.
  • The scheme supplies a stochastic variance-reduction route for non-finite-sum losses, a gap that classical SVR methods cannot fill.
  • Because matching costs are independent of minibatch size and only two assignments are solved, the method is among the fastest variance-reduction samplers, making it practical for large datasets.
  • The variance reduction is most pronounced at small minibatch sizes (k in the single digits), which is the regime where MMD/CORAL minibatch estimates are noisiest.
  • Periodic rematching as features evolve lets the same machinery track the changing distributions during training.

Where Pith is reading between the lines

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

  • The pairing logic is not tied to MMD/CORAL: any discrepancy estimator whose conditional mean given a subset is computable in closed form could be paired the same way, so contrastive or energy-based losses are natural testbeds.
  • The paper stops at pairs and quadruplets; iterating the assignment to build larger tuples would add cross-pair variance terms that the current costs ignore, so the observed shallow variance curve at large k may be improvable rather than fundamental.
  • Because PSDA changes the sampling distribution but does not add a control variate, combining it with unbiased SVR corrections could double-count variance reduction; a clean test would measure whether additive gains actually occur in practice.
  • The reported accuracy gains appear only under per-sampler hyperparameter tuning; a practical open question is how sensitive the method is when the minibatch size and UDA trade-off parameter are fixed across samplers.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. PSDA is a minibatch sampling scheme for UDA that constructs source-target pairs (Stage 1) and then pairs these pairs into quadruplets (Stage 2), with the aim of reducing the variance of stochastic MMD/CORAL discrepancy estimators. The pairings are obtained by solving linear assignment problems with closed-form cost matrices derived from conditional expectations of the discrepancy estimator. The paper reports Monte Carlo variance comparisons and image/acoustic classification experiments on Spawrious, Office-Home, and Humpbacks, where PSDA (especially Double-paired) is competitive with or better than several variance-reduced samplers and modern adversarial UDA methods, and at lower wall-clock cost than ORDERED.

Significance. The proposed mechanism is attractive and potentially useful: it is a drop-in sampler, does not modify the loss or architecture, and the cost matrices are cheap to compute. A correct proof that the pairings minimize expected gradient variance would be a real contribution to SVR for non-finite-sum losses. The empirical study is broad, with multiple baselines, three datasets, five repeats, and both MMD and CORAL. However, the central theoretical claim is not currently supported: the implemented costs are conditional-bias terms, not conditional MSEs, and the minibatch variance is not shown to be additive over the matched pairs. Because the paper's motivation and title rest on this claim, the theory must be repaired or explicitly downgraded before publication.

major comments (4)
  1. [§2.2.3, Eqs. (3), (10), (12)] The costs defined in Eq. (3) and Eq. (8) are conditional MSEs E[||\hat D-D||^2 | ...], but Eq. (10) computes ||E[\hat D | ...] - D||^2, the squared conditional bias. By the law of total variance, the omitted term E[||\hat D - E[\hat D | ...]||^2 | ...] is positive for the k>1 minibatch sizes used in the paper. The same substitution occurs in Eq. (12). Therefore the assignment (4) does not minimize the expected gradient variance claimed in the Abstract and §2.2; it minimizes a lower-bound proxy. This is the load-bearing gap in the paper.
  2. [§2.2.3, Eq. (9) (and analogously Eq. (11))] The conditional mean of \hat \mu_s given i∈B_s uses (k-1)\mu_s for the remaining points. Conditioned on i, the other k-1 points are sampled without replacement from the n_s-1 remaining examples, so the exact expectation is (φ(z_i)+(k-1)(n_s \mu_s-φ(z_i))/(n_s-1))/k. The O(1/n) correction is dropped in a derivation that presents the costs as exact. State this approximation and assess its effect on the assignment solution.
  3. [§2.2.1–§2.2.2 and §3.1] The step from the law of total expectation to a 'linear problem' is not valid for minibatches containing several matched pairs. The minibatch variance is not the sum of the per-pair costs in Eq. (4) or Eq. (8): covariance terms between co-occurring pairs contribute, and Stage-2 pairing changes the joint distribution of the remaining samples. No argument shows that the two-stage assignment optimizes the full minibatch objective, so the global-optimality claim in §3.1 is unsupported.
  4. [§2.2.3, Eqs. (13)–(14)] The CORAL cost matrices are asserted without derivation. They are not the exact conditional expectations of a sample covariance under without-replacement sampling; for example, when k=n_s Eq. (13) does not reduce to the full-sample covariance for a fixed included index i. Because the CORAL matching costs are built on these formulas, the CORAL variance-reduction claim is not established. If these are approximations, they should be stated and analysed.
minor comments (5)
  1. [§3.2, Tables 1–3] 'PSDA consistently achieves the highest or second-highest average accuracy' is not true for the Paired (Stage-1-only) variant. In Table 2 (MMD), Paired (45.4) ranks behind ORDERED (46.4), Double-paired (46.3), DPP (45.9), and VaRDASS (45.7); in Table 3 (MMD), Paired (90.9) is behind VaRDASS (92.0), ORDERED (91.8), and Double-paired (92.1). Please qualify.
  2. [§2.2.3] The update schedule for the matchings is not specified. State the rematching period used in the reported experiments and whether it is included in the wall-clock times of Table 4.
  3. [§2.2.1, Eq. (5)] '0≤Uij≤1' followed by 'U∈{0,1}' is redundant; state the integrality constraint once.
  4. [§3.1] The Monte Carlo variance plot has no error bars or number of seeds; please report MC repetitions to assess the 'three orders of magnitude' claim.
  5. [References] References to prior work by the same author (Napoli & White 2026a,b; Napoli 2026) are central baselines; if they are not yet published, provide sufficient detail or URLs so the comparisons can be verified. Also, the sampling distribution for k is typeset as 'k∼2 Uniform(3,7)'; this should presumably be k∼2^U(3,7).

Circularity Check

0 steps flagged

No significant circularity: PSDA's pairing costs are parameter-free closed-form expressions, and variance/accuracy claims are checked against external baselines and benchmarks.

full rationale

The paper's central derivation is self-contained: the cost matrices (Eqs. 10–14) are closed-form expressions in the feature embeddings and full-dataset statistics, with no fitted parameters that are then re-predicted. The variance-reduction claim is tested by independent Monte Carlo simulation against non-self baselines (Fig. 1), and accuracy is evaluated on external benchmarks (Spawrious, Office-Home, Humpbacks). Self-citations to prior work by the same author appear in the introduction and related work, but they are used for context and baselines, not as load-bearing justification for the assignment formulation. The derivation does contain a mathematical gap at §2.2.3: Eq. 10 computes ||E[\hat D|pair]−D||², the squared conditional bias, rather than the conditional MSE E[||\hat D−D||²|pair] defined in Eq. 3, thereby omitting the conditional variance for k>2. This is a correctness concern that may invalidate the theoretical optimality claim, but it is not circular: the paper does not define the pairing objective in terms of the result, nor does it fit a parameter to the target claim. No step reduces to its own input by construction or via a self-citation chain, so the appropriate circularity score is 0.

Axiom & Free-Parameter Ledger

3 free parameters · 6 axioms · 0 invented entities

Free parameters: the usual training hyperparameters, tuned per sampler by random search (10 draws) — a fairness-sensitive choice; the matching-update period, never reported; and the fixed RBF bandwidth set, from prior literature. Axioms: standard RKHS mathematics and LP integrality; two paper-specific approximations (full-mean substitution in Eq. 9; conditional-variance drop in Eqs. 10/12); and the unproven additivity of the two-stage matching objective. Invented entities: none — the quadruplet is an algorithmic construct, not a postulated model entity.

free parameters (3)
  • Minibatch size k and task trade-off λ, learning rate, weight decay = random search of 10 per sampler; k ~ 2^Uniform(3,7)
    Tuned independently per sampler on in-distribution validation; accuracy rankings depend on this budget choice.
  • Matching update (rematch) period = not reported
    Periodic recomputation of matchings is required (§2.2.3) but the frequency is never stated; a free implementation choice affecting cost and accuracy.
  • RBF mixture bandwidth set G = {0.001, 0.01, 0.1, 1, 10}
    Fixed kernel set from Li et al. (2018); standard, but chosen by hand.
axioms (6)
  • standard math RKHS: positive-definite kernel κ with reproducing property gives MMD the form ||µs−µt||²_H (Eq. 1).
    §2.1; standard functional-analytic background.
  • domain assumption Conditional expectation of the batch mean given i∈B_s equals [φ(zs_i)+(k−1)µs]/k (Eq. 9), i.e., leave-one-out mean ≈ full mean.
    Exactly sampling-with-replacement; approximate O(k/n) for without-replacement sampling used in practice.
  • ad hoc to paper E[||ˆD−D||²|i,j] ≈ ||E[ˆD−D|i,j]||² (conditional variance term drop).
    Eqs. 10/12 compute only the squared conditional mean; the omitted term varies at O(1/n) with the excluded example.
  • ad hoc to paper The two-stage matching objective aggregates linearly (no cross-pair terms when k≥2).
    §2.2.1 justifies linearity by the law of total expectation, but matching changes the co-occurrence distribution; global variance-optimality is never proven.
  • standard math Transportation-polytope integrality: the LP (4)-(7) with 0-1 bounds has integer optima.
    Needed for 'reduces to a linear assignment problem'; standard total unimodularity, unstated.
  • domain assumption DomainBed protocol with validation-based hyperparameter search is a sound measure of target accuracy.
    §3.2; standard practice but sensitive to search budget and seed selection.

pith-pipeline@v1.3.0-alltime-deepseek · 12218 in / 29186 out tokens · 231087 ms · 2026-08-01T10:02:52.387030+00:00 · methodology

0 comments
read the original abstract

Correlation alignment and the maximum mean discrepancy are two widely used distribution-matching frameworks for unsupervised domain adaptation (UDA). However, high variance in these losses has been shown to undermine their effectiveness in minibatch optimisation settings. Furthermore, the losses lack finite-sum structure, which renders them incompatible with classical stochastic variance reduction (SVR) methods. This paper proposes Paired Sampling for Domain Adaptation (PSDA), a novel SVR technique tailored to such objectives. PSDA pairs observations both within and across domains, to form quadruplets that are always sampled together during training. The pairings are designed to minimise expected gradient variance, and reduce to solving a set of linear assignment problems. Our simulations demonstrate reduced variance compared to related methods, and experiments on three domain shift datasets show improved target domain accuracy.

Figures

Figures reproduced from arXiv: 2607.20367 by Andrea Napoli.

Figure 1
Figure 1. Figure 1: Estimator variance vs k for six sampling algorithms. 3.1 Estimator variance We use Monte Carlo simulations to compare the estimator variance for different SVR methods across different values of k ( [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗

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

28 extracted references · 16 linked inside Pith

  1. [5]

    doi: 10.1109/CVPR46437.2021.01411

    ISSN 10636919. doi: 10.1109/CVPR46437.2021.01411. URLhttps: //arxiv.org/abs/2103.15796v2. Tianfan Fu and Zhihua Zhang. CPSG-MCMC: Clustering-Based Preprocessing method for Stochastic Gradient MCMC.AISTATS, pp. 841–850, 4

    arXiv 2021

  2. [7]

    Deep Residual Learning for Image Recognition.Pro- ceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2016-December: 770–778, 12

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep Residual Learning for Image Recognition.Pro- ceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2016-December: 770–778, 12

    2016

  3. [10]

    URLhttps://link.springer.com/article/10.1007/s10994-009-5152-4

    1007/S10994-009-5152-4/METRICS. URLhttps://link.springer.com/article/10.1007/s10994-009-5152-4. Yoshua Bengio, Jérome Louradour, Ronan Collobert, and Jason Weston. Curriculum learning.ICML, 382,

    doi:10.1007/s10994-009-5152-4

  4. [12]

    Weijie Liu, Hui Qian, Chao Zhang, Zebang Shen, Jiahao Xie, and Nenggan Zheng

    URLhttps://arxiv.org/abs/1810.03124v1. Weijie Liu, Hui Qian, Chao Zhang, Zebang Shen, Jiahao Xie, and Nenggan Zheng. Accelerating Stratified Sampling SGD by Reconstructing Strata.IJCAI,

    Pith/arXiv arXiv

  5. [13]

    URL https://proceedings.mlr.press/v37/long15.html

    PMLR. URL https://proceedings.mlr.press/v37/long15.html. Mingsheng Long, Zhangjie Cao, Jianmin Wang, and Michael I. Jordan. Conditional Adversarial Domain Adaptation. Advances in Neural Information Processing Systems, 2018-December:1640–1650, 5

    2018

  6. [14]

    URL https://arxiv.org/abs/1705.10667v4

    ISSN 10495258. URL https://arxiv.org/abs/1705.10667v4. Yucheng Lu, Youngsuk Park, Lifan Chen, Yuyang Wang, Christopher De Sa, and Dean Foster. Variance Reduced Training with Stratified Sampling for Forecasting Models.ICML, 139:7145–7155, 3

    Pith/arXiv arXiv

  7. [15]

    URL https://arxiv.org/abs/2103.02062v2

    ISSN 26403498. URL https://arxiv.org/abs/2103.02062v2. Aengus Lynch, Gbètondji J-S Dovonon, Jean Kaddour, and Ricardo Silva. Spawrious: A Benchmark for Fine Control of Spurious Correlation Biases.arXiv, 3

    Pith/arXiv arXiv

  8. [17]

    Diversity-Based Sampling for Imbalanced Domain Adaptation.EUSIPCO, 2024a

    Andrea Napoli and Paul White. Diversity-Based Sampling for Imbalanced Domain Adaptation.EUSIPCO, 2024a. Andrea Napoli and Paul White. Improving Domain Generalisation with Diversity-based Sampling.DCASE, 2024b. URLhttp://arxiv.org/abs/2410.04235. Andrea Napoli and Paul White. Order Matters: Improving Domain Adaptation by Reordering Data.arXiv, 2026a. Andre...

    Pith/arXiv arXiv

  9. [18]

    doi: 10.1037/met0000301

    ISSN 1082989X. doi: 10.1037/met0000301. URLhttps://pubmed.ncbi.nlm.nih. gov/32567870/. Harsh Rangwani, Sumukh K Aithal, Mayank Mishra, Arihant Jain, and R Venkatesh Babu. A Closer Look at Smooth- ness in Domain Adversarial Training.Proceedings of the 39th International Conference on Machine Learning,

    arXiv

  10. [20]

    9 Vladimir Vapnik.Statistical Learning Theory

    URLhttps://arxiv.org/abs/1412.3474v1. 9 Vladimir Vapnik.Statistical Learning Theory. Wiley, New York, US,

    Pith/arXiv arXiv

  11. [22]

    URLhttps://arxiv

    doi: 10.1109/CVPR.2017.572. URLhttps://arxiv. org/abs/1706.07522v1. Zirui Wang, Zihang Dai, Barnabas Poczos, and Jaime Carbonell. Characterizing and Avoiding Negative Transfer. CVPR, 2019-June:11285–11294, 11

    Pith/arXiv arXiv 2017

  12. [25]

    doi: 10.1609/AAAI.V33I01.33015741

    ISSN 2374-3468. doi: 10.1609/AAAI.V33I01.33015741. URL https://ojs.aaai.org/index.php/AAAI/article/view/4520. Marvin Zhang, Henrik Marklund, Nikita Dhawan, Abhishek Gupta, Sergey Levine, and Chelsea Finn. Adaptive Risk Minimization: Learning to Adapt to Domain Shift.Advances in Neural Information Processing Systems, 28: 23664–23678, 7

    doi:10.1609/aaai.v33i01.33015741

  13. [26]

    URLhttps://arxiv.org/abs/2007.02931v4

    ISSN 10495258. URLhttps://arxiv.org/abs/2007.02931v4. YiFan Zhang, Xue Wang, Jian Liang, Zhang Zhang, Liang Wang, Rong Jin, and Tieniu Tan. Free Lunch for Domain Adversarial Training: Environment Label Smoothing.ICLR, 2

    Pith/arXiv arXiv 2007

  14. [27]

    Peilin Zhao and Tong Zhang

    URLhttp://arxiv.org/abs/2302.00194. Peilin Zhao and Tong Zhang. Accelerating Minibatch Stochastic Gradient Descent using Stratified Sampling.arXiv, 5

    Pith/arXiv arXiv

  15. [28]

    URLhttps://arxiv.org/abs/1405.3080v1. 10

    Pith/arXiv arXiv

  16. [1962]

    doi: 10.1080/00401706.1962.10490022

    ISSN 15372723. doi: 10.1080/00401706.1962.10490022. Lilei Wu and Jie Liu. Contrastive Learning with Diverse Samples.Frontiers in Artificial Intelligence and Applications, 372:2672–2679, 9

    arXiv 1962

  17. [1998]

    Deep Hashing Network for Unsupervised Domain Adaptation.CVPR 2017,

    Hemanth Venkateswara, Jose Eusebio, Shayok Chakraborty, and Sethuraman Panchanathan. Deep Hashing Network for Unsupervised Domain Adaptation.CVPR 2017,

    2017

  18. [2009]

    URLhttps://dl.acm.org/doi/10.1145/1553374.1553380

    doi: 10.1145/1553374.1553380. URLhttps://dl.acm.org/doi/10.1145/1553374.1553380. Abhimanyu Dubey, Vignesh Ramanathan, Alex Pentland, and Dhruv Mahajan. Adaptive Methods for Real-World Domain Generalization.CVPR,

    arXiv

  19. [2014]

    URLhttps://arxiv.org/abs/1412.6980v9

    doi: 10.48550/arxiv.1412.6980. URLhttps://arxiv.org/abs/1412.6980v9. 8 Pang Wei Koh, Shiori Sagawa, Henrik Marklund, Sang Michael Xie, Marvin Zhang, Akshay Balsubramani, Weihua Hu, Michihiro Yasunaga, Richard Lanas Phillips, Irena Gao, Tony Lee, Etienne David, Ian Stavness, Wei Guo, Berton A. Earnshaw, Imran S. Haque, Sara Beery, Jure Leskovec, Anshul Kun...

    Pith/arXiv arXiv doi:10.48550/arxiv.1412.6980

  20. [2015]

    doi: 10.48550/arxiv.1512.03385

    ISSN 10636919. doi: 10.48550/arxiv.1512.03385. URLhttps://arxiv.org/abs/1512.03385v1. Wenyu Jiang, Hao Cheng, MingCai Chen, Chongjun Wang, and Hongxin Wei. DOS: Diverse Outlier Sampling for Out-of-Distribution Detection.ICLR,

    Pith/arXiv arXiv doi:10.48550/arxiv.1512.03385

  21. [2016]

    URLhttps://arxiv.org/abs/1607.01719v1

    ISSN 16113349. URLhttps://arxiv.org/abs/1607.01719v1. Eric Tzeng, Judy Hoffman, Ning Zhang, Kate Saenko, and Trevor Darrell. Deep Domain Confusion: Maximizing for Domain Invariance.arXiv, 12

    Pith/arXiv arXiv

  22. [2017]

    URLhttps://proceedings.mlr.press/v54/fu17a

    ISSN 2640-3498. URLhttps://proceedings.mlr.press/v54/fu17a. html. 7 Table 3: Average test accuracy for Humpbacks by data split. Method Domain 1 Domain 2 Domain 3 Domain 4 A verage ERM 70.3±2.7 92.0±1.9 78.1±3.0 96.2±0.6 84.2±1.1 DANN 60.5±4.2 90.1±2.1 63.9±6.1 76.1±11.1 72.6±3.4 CDAN 61.6±4.2 82.2±4.7 73.5±3.0 84.4±0.6 75.4±1.7 CDAN + SDAT 63.7±3.0 81.2±6...

    1944

  23. [2018]

    doi: 10.1109/CVPR.2018.00566

    ISSN 10636919. doi: 10.1109/CVPR.2018.00566. Jingchang Liu and Linli Xu. Accelerating Stochastic Gradient Descent Using Antithetic Sampling.arXiv, 10

    arXiv 2018

  24. [2019]

    doi: 10.1109/CVPR.2019.01155

    ISSN 10636919. doi: 10.1109/CVPR.2019.01155. URLhttps://arxiv. org/abs/1811.09751v4. B. P. Welford. Note on a Method for Calculating Corrected Sums of Squares and Products.Technometrics, 4(3): 419–420,

    arXiv 2019

  25. [2020]

    doi: 10.1007/978-3-030-58589-1{\_}28

    ISSN 16113349. doi: 10.1007/978-3-030-58589-1{\_}28. URL https://arxiv.org/pdf/1912.03699. Diederik P. Kingma and Jimmy Lei Ba. Adam: A Method for Stochastic Optimization.3rd International Conference on Learning Representations, ICLR 2015 - Conference Track Proceedings, 12

    Pith/arXiv arXiv 1912

  26. [2021]

    URL https://arxiv.org/abs/2112.06007v1

    ISSN 10495258. URL https://arxiv.org/abs/2112.06007v1. Philipp Baumann, Olivier Goldschmidt, Dorit S. Hochbaum, and Jason Yang. A Fast and Effective Method for Euclidean Anticlustering: The Assignment-Based-Anticlustering Algorithm.arXiv, 1

    Pith/arXiv arXiv

  27. [2023]

    Andrea Napoli

    URLhttps://arxiv.org/abs/2303.05470v3. Andrea Napoli. Online Variance Reduction for Domain Adaptation on Streaming Data.arXiv,

    Pith/arXiv arXiv

  28. [2026]

    org/abs/2601.06351

    URLhttp://arxiv. org/abs/2601.06351. Shai Ben-David, John Blitzer, Koby Crammer, and Fernando Pereira. Analysis of Representations for Domain Adaptation.NeurIPS, 19,

    Pith/arXiv arXiv