Pith. sign in

REVIEW 3 major objections 5 minor 42 references

Towards Fair and Privacy-Aware Transfer Learning for Educational Predictive Modeling: A Case Study on Retention Prediction in Community Colleges

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Pith's one-line read Public institution data can predict when a transfer model will fail.

desk verdict Large-scale, credible direct-transfer results undercut by an in-sample contextual-similarity 'forecast' that needs out-of-sample validation. read the letter →

arxiv 2501.06913 v1 pith:VOBN3K4J submitted 2025-01-12 cs.CY

classification cs.CY
keywords PredictiveAnalyticsTransferLearningAlgorithmicFairnessPrivacyIntersectionalityCollegeRetentionCommunityCollegesHigherEducation
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

This paper tries to establish that community colleges, which often cannot build their own retention-prediction models or access other institutions' student data, can still transfer models responsibly using only public institutional information. It argues that a simple contextual-similarity score built from public IPEDS data predicts how much a source model's performance will drop at a target institution, and that this score can guide model selection. It also claims that a developer-side sequential-training procedure improves fairness without sacrificing accuracy, and that user-side customized evaluation thresholds outperform established source-free adaptation methods. If right, resource-poor institutions could get better, fairer predictive models without sharing private student records.

What carries the argument

The load-bearing object is the contextual similarity score, computed with Gower's distance across six categories of public IPEDS institutional data and combined into one overall score using a 'Coefficient-Significance Weighted Similarity' weight $W_i = |\beta_i|/(1+p_i)$, where $\beta_i$ and $p_i$ come from a regression of AUC Drop on the six category similarities. This score powers the Most Similar Training Institution selection rule and the 0.6 similarity threshold used in sequential training. A second mechanism is the evaluation-threshold strategy: instead of adapting the model, the target institution picks decision thresholds from its own historical outcome rates, either overall or per demographic group, which changes specificity and fairness without retraining.

What would settle it

Compute the weighted contextual similarity for a held-out set of source–target institution pairs that were not used to fit the AUC Drop regression, and compare the predicted AUC drop with the observed drop; if the correlation is near zero or negative, the forecasting claim fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that direct cross-institutional transfer of retention models degrades both performance and fairness, but the performance loss is largely predictable from publicly available contextual information. Regressing AUC Drop on six institutional similarity categories yields an R-squared of 0.600, and an overall similarity score built from the significant coefficients separates institution pairs that transfer well from those that do not. Fairness change, measured as the AUC Gap, is not predictable from the same variables (R-squared 0.056). For developers, sequentially training on the most similar institution and then on a demographically dissimilar institution with similarity above 0.6 improves the AUC Gap without changing test-AUC distributions. For users, group-optimal evaluation thresholds improve specificity and equalized odds relative to default and overall-optimal thresholds, while source-free domain adaptation methods (SHOT, TENT, pseudo-labeling) do not help and often hurt.

Load-bearing premise

The whole forecasting idea depends on the assumption that the same contextual factors that predicted transfer success for the institutions in this study will also predict it for colleges not in the study, since the weights were estimated on the same pairs used to show the pattern.

Editorial extensions

If this is right

  • A community college can use only public IPEDS data to rank candidate source institutions and avoid the worst-performing transfer models, without access to any other institution's student records.
  • Model developers can build fairer transfer models under privacy constraints by sequentially training on the most contextually similar institution plus a demographically dissimilar one, provided initial unfairness is high enough.
  • Target institutions with no local labeled data and no model details can improve deployed-model specificity and group fairness simply by choosing group-specific evaluation thresholds.
  • Source-free domain adaptation methods that work in computer vision should not be assumed to work for educational retention transfer; in this setting they underperform direct transfer.
  • Fairness degradation across institutions is largely unpredictable from contextual similarity, so fairness must be monitored locally rather than forecast from public data.

Reading between the lines

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

  • The forecasting claim would be strengthened by an out-of-sample test: fitting the AUC Drop regression on one subset of institution pairs and checking predictions on held-out pairs. The paper's R-squared of 0.600 is computed in-sample, so a reader should treat the 'forecast' language as a hypothesis until that test is run.
  • The same weighted-context recipe could be tested for other transfer tasks, such as course-level performance prediction or dropout prediction in four-year institutions, where the relevant contextual categories may differ.
  • Group-optimal thresholding is a zero-data-sharing, zero-retraining intervention, so it could be applied immediately by any institution that already deploys a vendor model, making it an attractive fairness remedy in practice.
  • Because the contextual weights were derived with cost similarity removed post hoc, future work could test whether the weighting scheme is stable across different regression specifications and institution samples.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper examines cross-institutional transfer learning for first-year retention prediction at U.S. community colleges, using administrative records from 4 research universities and 23 community colleges covering over 800,000 students across 7 cohorts. It evaluates three sets of strategies: direct transfer of pre-trained models, sequential training with EWC, and source-free domain adaptation plus customized evaluation thresholds. The main empirical findings are that direct transfer causes both performance and fairness degradation relative to ideal local models; that a contextual similarity score computed from IPEDS data correlates with the performance drop (R-squared 0.600); that sequential training selecting a demographically dissimilar second institution can improve fairness without significantly changing test AUC distributions; and that group-optimal thresholds improve specificity and equalized odds without harming MCC. The paper argues that publicly available contextual information can forecast transfer performance drops and guide model selection under privacy constraints.

Significance. If the forecasting claim holds, this would be a practically valuable result for resource-constrained institutions choosing among externally trained models. The study is notable for its large, multi-institutional dataset, its focus on community colleges, and its attention to fairness and privacy. The direct-transfer degradation result is measured on held-out 2019 students and appears credible. The code repository is a concrete reproducibility asset. However, the central forecasting claim currently rests on an in-sample regression, so the significance of the paper hinges on whether that claim is re-validated out of sample.

major comments (3)
  1. [4.4.3 and 5.1] The contextual similarity weights W_i = |beta_i|/(1+p_i) are estimated from a regression of AUC Drop on the six category-level similarity scores, and the same fitted weights are then used to construct the overall similarity score whose association with AUC Drop is reported (R-squared 0.600, Figure 3). This is an in-sample fit, so the claim in the abstract and Section 5.1 that 'publicly available contextual information can forecast these performance drops' is not supported by the presented analysis. The paper should provide an out-of-sample evaluation (for example, leave-one-institution-out cross-validation or a held-out set of institution pairs), report the number of observations, adjusted R-squared, and prediction error on held-out pairs, or reframe the claim as descriptive rather than predictive.
  2. [5.1 and 5.2] The post hoc removal of cost similarity from the overall similarity score (Section 5.1) and the choice of the 0.6 similarity threshold for selecting Training Institution 2 (Section 5.2) are additional model-selection decisions made on the same data used to demonstrate their benefits. Because these decisions are not justified by a principled criterion or validated on independent data, the reported improvements for MSTI and sequential training may be optimistic. A sensitivity analysis or a pre-specified decision rule would help address this concern.
  3. [5.2] The claim that sequential training improves fairness without compromising performance is based on small and selected subsets: Figure 7 reports results for only 11 target institutions, and the WTNDD test for Test AUC differences is a null result that does not quantify the possible performance trade-off. The paper should report effect sizes and confidence intervals for the change in AUC Gap and Test AUC, and should clarify how the 11 institutions were selected relative to the full set.
minor comments (5)
  1. [4.4.1] The formula for AUC Gap is not defined clearly; the notation ED_k and the structure of the maximum over groups are ambiguous. Please provide a precise definition with all terms explained.
  2. [5.3 and Figure 8] The claim that SFDA methods underperform compared to direct transfer is based on visual comparison of boxplots; include paired statistical tests across target institutions (e.g., Wilcoxon signed-rank test) to support the claim.
  3. [4.4.2] The WTNDD threshold of 5% of the data range is arbitrary; please provide a justification or a sensitivity analysis showing the robustness of the conclusions to this threshold.
  4. [Figure 3] The caption uses the term 'Inverted AUC drop' without defining it; please define it in the text or caption so readers know it is the negative of AUC Drop.
  5. [Global] The repeated 'Manuscript submitted to ACM' footer appears on many pages; this production artifact should be removed in the final version.

Circularity Check

1 steps flagged · score 6.0 of 10

The contextual-similarity 'forecast' is built from the same regression it is later used to explain, so the central portability prediction is an in-sample fit rather than a validated forecast.

  1. fitted input called prediction [Section 4.4.3 (Eq. for W_i) and Section 5.1 (R-squared and Figure 3)]
    "To construct an overall similarity score between source and target institutions, we utilize the results of a regression analysis on AUC Drop (ΔAUC(local,·)) (see Section 5.1 for more details). ... The weight for each similarity measure i is calculated as: W_i = |β_i| / (1 + p_i) ... The regression model for AUC Drop has an R-squared value of 0.600 ... Following the regression results for AUC Drop, we calculate an overall similarity score ... Figure 3 illustrates the relationship between overall similarity and three key metrics: Test AUC, Inverted AUC drop, and AUC Gap."

    The overall similarity score is constructed from the coefficients and p-values of a regression whose target is exactly the AUC Drop that the score is later claimed to forecast. Because the same source-target pairs that produced β_i and p_i are the ones plotted in Figure 3 and summarized by R² = 0.600, the association is in-sample by construction: the composite predictor was tuned to the outcome on this dataset. The additional decision to drop cost similarity because it 'tends to enlarge the AUC drop' is further in-sample model selection on the same pairs.

full rationale

The paper's non-circular contributions include the direct-transfer degradation results (Figure 1) and the strategy comparisons for sequential training and customized thresholds, which are evaluated against local baselines without relying on the fitted contextual weights. However, the headline forecast claim rests on the Coefficient-Significance Weighted Similarity Method: W_i is defined from the regression coefficients and p-values of the AUC Drop regression, and Figure 3 then displays the relationship between the resulting overall similarity and the same AUC Drop on the same source-target pairs. This is an in-sample fit, not a predictive validation. The fairness regression's near-zero R² (0.056) reinforces that only the fitted performance-drop relationship carries the forecast claim. Because the central 'forecast' claim reduces to a fitted composite of the same data, the circularity score is 6; the remaining transfer-learning strategies are independent and would not warrant a high score on their own.

Assumptions & free parameters 3 free parameters · 4 assumptions · 2 invented entities

The central claims rest on the feature schema, IPEDS context measures, fairness metrics, and the temporal split; all are domain assumptions rather than derived results. The contextual similarity weights and the WTNDD threshold are quantities introduced by the authors without independent validation.

free parameters (3)
  • Contextual similarity regression weights beta_i = not reported
    Weights for each of six contextual categories are fitted by regressing AUC Drop on similarity scores across all source-target pairs (Sections 4.4.3 and 5.1). These weights define the overall similarity score used for MSTI selection and for the forecasting claim.
  • Demographic dissimilarity selection threshold 0.6 for Training Institution 2 = 0.6
    Chosen in Section 5.2 'to avoid performance drops' based on the same data, then used in sequential training; no independent validation.
  • WTNDD threshold = 5% of data range
    The Wasserstein Threshold for Notable Distribution Difference is set at 5% of the data range (Section 4.4.2) by convention, and all distribution comparisons pass or fail against it.
assumptions (4)
  • domain assumption The shared schema of common variables across all 27 institutions captures the predictors needed for retention prediction.
    Used in Section 4.1.1 to create a common feature set; if important local indicators are missing, transfer results are affected.
  • domain assumption IPEDS contextual factors (64 variables across six categories) adequately operationalize institutional context relevant to model transfer.
    The entire contextual similarity analysis in Sections 4.4.3 and 5.1 depends on this.
  • domain assumption AUC Gap and Equalized Odds are appropriate fairness criteria for this setting.
    Fairness conclusions in RQ1 through RQ3 rely on these metrics; alternative metrics could change conclusions.
  • domain assumption Students entering Fall 2013 through Fall 2018 (training) and Fall 2019 (test) are exchangeable enough for evaluation.
    The temporal split assumes no distribution shift beyond observed years; cohorts after 2019 are excluded due to COVID.
invented entities (2)
  • Coefficient-Significance Weighted Similarity Method
    purpose: Produces an overall contextual similarity score from six category similarities by weighting with regression coefficients and p-values (Section 4.4.3).
    This is a new weighting scheme introduced for this paper; it is not benchmarked against alternative weighting schemes and is fit on the same data.
  • Wasserstein Threshold for Notable Distribution Difference (WTNDD)
    purpose: Declares whether two performance distributions differ meaningfully, using a 5% of range threshold (Section 4.4.2).
    A new threshold criterion without external validation; all conclusions about distribution differences depend on this arbitrary cutoff.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards Fair and Privacy-Aware Transfer Learning for Educational Predictive Modeling: A Case Study on Retention Prediction in Community Colleges." pith.science (2026). https://pith.science/paper/VOBN3K4J

@misc{pith2026250106913,
  author       = {Pith},
  title        = {Pith review of: Towards Fair and Privacy-Aware Transfer Learning for Educational Predictive Modeling: A Case Study on Retention Prediction in Community Colleges},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VOBN3K4J}},
  note         = {Machine review of arXiv:2501.06913}
}
read the original abstract

Predictive analytics is widely used in learning analytics, but many resource-constrained institutions lack the capacity to develop their own models or rely on proprietary ones trained in different contexts with little transparency. Transfer learning holds promise for expanding equitable access to predictive analytics but remains underexplored due to legal and technical constraints. This paper examines transfer learning strategies for retention prediction at U.S. two-year community colleges. We envision a scenario where community colleges collaborate with each other and four-year universities to develop retention prediction models under privacy constraints and evaluate risks and improvement strategies of cross-institutional model transfer. Using administrative records from 4 research universities and 23 community colleges covering over 800,000 students across 7 cohorts, we identify performance and fairness degradation when external models are deployed locally without adaptation. Publicly available contextual information can forecast these performance drops and offer early guidance for model portability. For developers under privacy regulations, sequential training selecting institutions based on demographic similarities enhances fairness without compromising performance. For institutions lacking local data to fine-tune source models, customizing evaluation thresholds for sensitive groups outperforms standard transfer techniques in improving performance and fairness. Our findings suggest the value of transfer learning for more accessible educational predictive modeling and call for judicious use of contextual information in model training, selection, and deployment to achieve reliable and equitable model transfer.

Figures

Figures reproduced from arXiv: 2501.06913 by the authors.

Figure 1
Figure 1. Test AUC (left) and AUC Gaps (right) of pre-trained models for each target institutions. Blue circles represent the tested metrics of pre-trained direct transfer models for each target institution, while red squares indicate the tested metrics of ideal local models. depending on the source institution [PITH_FULL_IMAGE:figures/full_fig_p009_1.png] view at source ↗
Figure 2
Figure 2. Association between contextual similarity metrics and AUC Drop (left) and AUC Gap (right). A negative coefficient indicates a reduction in the AUC Drop or AUC Gap. Error bars represent 95% confidence intervals. perform better, with smaller AUC drops, when the source and target institutions are more similar in terms of school, academic, demographic, and completion factors [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Relationship between overall contextual similarity and the model’s transferability and fairness. Blue circles represent the tested metric of pre-trained direct transfer models on each target institution. relationship between overall similarity and three key metrics: Test AUC, Inverted AUC drop, and AUC Gap. The left two plots demonstrate a positive linear trend, indicating that higher overall similarity between inst… view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Performance distributions across all target institutions in the dataset, comparing Test AUC for three scenarios: MSTI model, the ideal local model, and the expected outcome. The histograms are normalized to have an area of 1 (density) [PITH_FULL_IMAGE:figures/full_fig…
Figure 5
Figure 5. Figure 5: Differences between MSTI’s Test AUC and AUC Gap with expected values across target institutions. respecting privacy constraints. First, for each target institution, we firstly identify the MSTI model. Next, among other possible training institutions with an overall sim…
Figure 6
Figure 6. Figure 6: Distribution of performance and fairness for MSTI model and sequential model. (a) Top: Test AUC distributions when the source is MSTI model and the sequential model under varying MSTI AUC gap thresholds (>=0.00, 0.05, and 0.10). (b) Bottom: AUC Gap distributions for MS…
Figure 7
Figure 7. Figure 7: Differences in AUC Gap and Test AUC between the MSTI model and the sequential model for each target institution when MSTI AUC Gap > 0.10. A positive value indicates an improvement of tested metric with sequential training. model. However, this reduction in AUC Gap is o…
Figure 8
Figure 8. Figure 8: Test AUC distributions across four different methods—Direct Transfer (blue), SHOT (orange), TENT (green), and Pseudo-Labeling (red)—for each target institution. Each box represents the interquartile range (IQR) of the AUC values, with the whiskers extending to the mini…
Figure 9
Figure 9. Figure 9 [PITH_FULL_IMAGE:figures/full_fig_p013_9.png]
Figure 10
Figure 10. Figure 10: Distribution of Equalized Odds (EO) scores of direct transfer models. (a) Top: EO scores for male vs. female under the usage of default threshold, group optimal threshold, and overall optimal threshold. (b) Bottom: EO scores for URM vs. Non-URM under the usage of same…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 36 canonical work pages

  1. [1]

    Fard, Ratna B

    Sattar Ameri, Mahtab J. Fard, Ratna B. Chinnam, and Chandan K. Reddy. 2016. Survival Analysis Based Framework for Early Prediction of Student Dropouts. In Proceedings of the 25th ACM International Conference on Information and Knowledge Management (CIKM ’16) . Association for Computing Machinery, 903–912

  2. [2]

    Lovenoor Aulck, Dev Nambi, Nishant Velagapudi, Joshua Blumenstock, and Jevin West. 2019. Mining University Registrar Records to Predict First-Year Undergraduate Attrition. In Proceedings of the 12th International Conference on Educational Data Mining (EDM 2019) . 9–18

  3. [3]

    Rosenthal

    Cédric Beaulac and Jeffrey S. Rosenthal. 2019. Predicting University Students’ Academic Success and Major Using Random Forests. Research in Higher Education 60, 7 (2019), 1048–1064

  4. [4]

    Johannes Berens, Kerstin Schneider, Simon Görtz, Simon Oster, and Julian Burghoff. 2019. Early Detection of Students at Risk: Predicting Student Dropouts Using Administrative Student Data from German Universities and Machine Learning Methods. Journal of Educational Data Mining 11, 3 (2019), 1–41

  5. [5]

    Bird, Benjamin L

    Kelli A. Bird, Benjamin L. Castleman, Zachary Mabel, and Yifeng Song. 2021. Bringing Transparency to Predictive Analytics: A Systematic Comparison of Predictive Modeling Methods in Higher Education. AERA Open 7 (2021), 233285842110376. https://doi.org/10.1177/23328584211037630

  6. [6]

    Bird, Benjamin L

    Kelli A. Bird, Benjamin L. Castleman, and Yifeng Song. 2024. Are algorithms biased in education? Exploring racial bias in predicting community college student success. Journal of Policy Analysis and Management (31 January 2024)

  7. [7]

    Varshney, Skyler Speakman, Zairah Mustahsan, and Supriyo Chakraborty

    Amanda Coston, Karthikeyan Natesan Ramamurthy, Dennis Wei, Kush R. Varshney, Skyler Speakman, Zairah Mustahsan, and Supriyo Chakraborty

  8. [8]

    A. Gardner. 2022. Persistence and Retention: Fall 2020 Beginning Postsecondary Student Cohort . Technical Report. National Student Clearinghouse Research Center, Herndon, VA

Show all 42 references
  1. [9]

    Joshua Gardner, Renzhe Yu, Quan Nguyen, Christopher Brooks, and Rene Kizilcec. 2023. Cross-Institutional Transfer Learning for Educational Models: Implications for Model Performance, Fairness, and Equity. In 2023 ACM Conference on Fairness, Accountability, and Transparency (FA...

  2. [10]

    Dragan Gašević, Shane Dawson, Tim Rogers, and Danijela Gasevic. 2016. Learning analytics should not promote one size fits all: The effects of instructional conditions in predicting academic success. The Internet and Higher Education 28 (2016), 68–84

  3. [12]

    J. C. Gower. 1971. A general coefficient of similarity and some of its properties. Biometrics 27, 4 (1971), 857–871

  4. [13]

    Hoi, Doyen Sahoo, Jing Lu, and Peilin Zhao

    Steven C.H. Hoi, Doyen Sahoo, Jing Lu, and Peilin Zhao. 2021. Online learning: A comprehensive survey. Neurocomput. 459, C (Oct. 2021), 249–289

  5. [14]

    Hassan Ismail Fawaz, Germain Forestier, Jonathan Weber, Lhassane Idoumghar, and Pierre-Alain Muller. 2018. Transfer learning for time series classification. In 2018 IEEE International Conference on Big Data (Big Data) . IEEE, 1367–1376. https://doi.org/10.1109/bigdata.2018.862...

  6. [15]

    Jing Jiang and ChengXiang Zhai. 2007. Instance Weighting for Domain Adaptation in NLP. In Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics , Annie Zaenen and Antal van den Bosch (Eds.). Association for Computational Linguistics, Prague, Cz...

  7. [16]

    Rusu, Kieran Milan, John Quan, Tiago Ramalho, Agnieszka Grabska-Barwinska, Demis Hassabis, Claudia Clopath, Dharshan Kumaran, and Raia Hadsell

    James Kirkpatrick, Razvan Pascanu, Neil Rabinowitz, Joel Veness, Guillaume Desjardins, Andrei A. Rusu, Kieran Milan, John Quan, Tiago Ramalho, Agnieszka Grabska-Barwinska, Demis Hassabis, Claudia Clopath, Dharshan Kumaran, and Raia Hadsell. 2017. Overcoming catastrophic forget...

  8. [17]

    Kizilcec and Hansol Lee

    René F. Kizilcec and Hansol Lee. 2022. Algorithmic Fairness in Education. In Ethics in Artificial Intelligence in Education . Routledge

  9. [18]

    intersectionally fair

    Youjin Kong. 2022. Are “intersectionally fair” ai algorithms really fair to women of color? A philosophical analysis. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency . Association for Computing Machinery, 485–494. https://doi.org/10.1145...

  10. [19]

    Moritz Laurer, Wouter van Atteveldt, Andreu Casas, and Kasper Welbers. 2024. Less Annotating, More Classifying: Addressing the Data Scarcity Issue of Supervised Machine Learning with Deep Transfer Learning and BERT-NLI. Political Analysis 32, 1 (2024), 84–100

  11. [20]

    Dong-Hyun Lee. 2013. Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks. In Workshop on Challenges in Representation Learning, ICML

  12. [21]

    Kizilcec

    Xingyu Li, Difan Song, Miaozhe Han, Yu Zhang, and Rene F. Kizilcec. 2021. On the limits of algorithmic prediction across the globe. arXiv:2103.15212 [cs.LG] https://arxiv.org/abs/2103.15212

  13. [22]

    Jian Liang, Dapeng Hu, and Jiashi Feng. 2020. Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation. In International Conference on Machine Learning (ICML)

  14. [23]

    Mariia Luzan and Christopher Brooks. 2024. Evaluation of the instance weighting strategy for transfer learning of educational predictive models. Proceedings of Machine Learning Research 257 (9 August 2024), 19–28

  15. [24]

    Debmalya Mandal, Samuel Deng, Suman Jana, and Daniel Hsu. 2020. Ensuring Fairness Beyond the Training Data. InAdvances in Neural Information Processing Systems (NeurIPS)

  16. [25]

    Ninareh Mehrabi, Fred Morstatter, Nripsuta Saxena, Kristina Lerman, and Aram Galstyan. 2021. A Survey on Bias and Fairness in Machine Learning. ACM Computing Surveys (CSUR) 54, 6 (2021), 1–35

  17. [26]

    Vanessa Smith Morest. 2013. From Access to Opportunity: The Evolving Social Roles of Community Colleges. The American Sociologist 44, 4 (2013), 319–328. http://www.jstor.org/stable/42635370

  18. [27]

    Abdallah Namoun and Abdullah Alshanqiti. 2021. Predicting Student Performance Using Data Mining and Learning Analytics Techniques: A Systematic Literature Review. Applied Sciences 11, 1 (2021), 237. https://doi.org/10.3390/app11010237

  19. [28]

    Jaclyn Ocumpaugh, Ryan Baker, Sujith Gowda, Neil Heffernan, and Cristina Heffernan. 2014. Population validity for educational data mining models: A case study in affect detection. British Journal of Educational Technology 45, 3 (May 2014), 487–501

  20. [29]

    Sinno Jialin Pan and Qiang Yang. 2010. A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering 22, 10 (2010), 1345–1359. https://doi.org/10.1109/TKDE.2009.191

  21. [30]

    Ashkan Rezaei, Anqi Liu, Omid Memarrast, and Brian D. Ziebart. 2021. Robust Fairness Under Covariate Shift. Proceedings of the AAAI Conference on Artificial Intelligence 35, 11 (May 2021), 9419–9427

  22. [31]

    Jessica Schrouff, Natalie Harris, Oluwasanmi O Koyejo, Ibrahim Alabdulmohsin, Eva Schnider, Krista Opsahl-Ong, Alexander Brown, Subhrajit Roy, Diana Mincu, Chrsitina Chen, Awa Dieng, Yuan Liu, Vivek Natarajan, Alan Karthikesalingam, Katherine A Heller, Silvia Chiappa, and Alex...

  23. [32]

    Reed Scull, Mark Andrew Perkins, Jonathan W

    W. Reed Scull, Mark Andrew Perkins, Jonathan W. Carrier, and Michael Barber. 2022. Community College Institutional Researchers’ Knowledge, Experience, and Perceptions of Machine Learning. Community College Journal of Research and Practice 47, 5 (2022), 354–368

  24. [33]

    Doug Shapiro, Afet Dundar, Faye Huie, Phoebe Khasiala Wakhungu, Ayesha Bhimdiwala, and Sean Eric Wilson. 2019. Completing College: A State-Level View of Student Completion Rates (Signature Report No. 16a) . Technical Report. National Student Clearinghouse Research Center, Herndon, VA

  25. [34]

    Vinitra Swamy, Mirko Marras, and Tanja Käser. 2022. Meta Transfer Learning for Early Success Prediction in MOOCs. In Proceedings of the Ninth ACM Conference on Learning @ Scale (New York City, NY, USA)(L@S ’22). Association for Computing Machinery, New York, NY, USA, 121–132

  26. [35]

    US Congress. 1974. Family Educational Rights and Privacy Act (FERPA). https://www.law.cornell.edu/uscode/text/20/1232g 20 U.S.C. § 1232g

  27. [36]

    Cédric Villani. 2008. Optimal Transport: Old and New . Grundlehren der mathematischen Wissenschaften, Vol. 338. Springer, Berlin, Heidelberg

  28. [37]

    Dequan Wang, Evan Shelhamer, Shaoteng Liu, Bruno Olshausen, and Trevor Darrell. 2021. Tent: Fully Test-Time Adaptation by Entropy Minimization. In International Conference on Learning Representations . https://openreview.net/forum?id=uXl3bZLkr3c

  29. [38]

    Zhen Xu, Joseph Olson, Nicole Pochinki, Zhijian Zheng, and Renzhe Yu. 2024. Contexts Matter but How? Course-Level Correlates of Performance and Fairness Shift in Predictive Model Transfer. In The 14th Learning Analytics and Knowledge Conference (LAK ’24) (Kyoto, Japan, March 1...

  30. [39]

    Kizilcec

    Renzhe Yu, Hansol Lee, and René F. Kizilcec. 2021. Should college dropout prediction models include protected attributes?. In Proceedings of the Eighth ACM Conference on Learning @ Scale . Association for Computing Machinery, New York, NY, USA, 91–100

  31. [40]

    Zhiqi Yu, Jingjing Li, Zhekai Du, Lei Zhu, and Heng Tao Shen. 2023. A Comprehensive Survey on Source-free Domain Adaptation. arXiv:2302.11803 [cs.LG] https://arxiv.org/abs/2302.11803 Manuscript submitted to ACM Towards Fair and Privacy-Aware Transfer Learning for Educational P...

  32. [41]

    Chuanting Zhang, Haixia Zhang, Jingping Qiao, Dongfeng Yuan, and Minggao Zhang. 2019. Deep Transfer Learning for Intelligent Cellular Traffic Prediction Based on Cross-Domain Big Data. IEEE Journal on Selected Areas in Communications 37, 6 (2019), 1389–1401

  33. [42]

    Hoi, Jialei Wang, and Bin Li

    Peilin Zhao, Steven C.H. Hoi, Jialei Wang, and Bin Li. 2014. Online Transfer Learning. Artificial Intelligence 216 (2014), 76–102. https://doi.org/10. 1016/j.artint.2014.06.003 Manuscript submitted to ACM

  34. [2019]

    In Proceedings of the 2019 AAAI/ACM Conference on AI, Ethics, and Society

    Fair Transfer Learning with Missing Protected Attributes. In Proceedings of the 2019 AAAI/ACM Conference on AI, Ethics, and Society . 91–98

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.