Regularity as Structural Amplifier, Not Trap: A Causal and Archetype-Based Analysis of Dropout in a Constrained Engineering Curriculum

• URL: http://arxiv.org/abs/2511.18979v1
• Date: Mon, 24 Nov 2025 10:53:41 GMT
• Title: Regularity as Structural Amplifier, Not Trap: A Causal and Archetype-Based Analysis of Dropout in a Constrained Engineering Curriculum
• Abstract summary: Engineering programmes, particularly in Latin America, are often governed by rigid curricula and strict regularity rules.<n>This study tests that causal hypothesis using the CAPIRE framework.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Engineering programmes, particularly in Latin America, are often governed by rigid curricula and strict regularity rules that are claimed to create a Regularity Trap for capable students. This study tests that causal hypothesis using the CAPIRE framework, a leakage-aware pipeline that integrates curriculum topology and causal estimation. Using longitudinal data from 1,343 civil engineering students in Argentina, we formalize academic lag (accumulated friction) as a treatment and academic velocity as an ability proxy. A manual LinearDML estimator is employed to assess the average (ATE) and conditional (CATE) causal effects of lag on subsequent dropout, controlling for macro shocks (strikes, inflation). Results confirm that academic lag significantly increases dropout risk overall (ATE = 0.0167, p < 0.0001). However, the effect decreases sharply for high-velocity (high-ability) students, contradicting the universal Trap hypothesis. Archetype analysis (UMAP/DBSCAN) shows that friction disproportionately harms trajectories already characterized by high initial friction and unstable progression. 8 We conclude that regularity rules function as a Structural Amplifier of pre-existing vulnerability rather than a universal trap. This has direct implications for engineering curriculum design, demanding targeted slack allocation and intervention policies to reduce friction at core basic-cycle courses

Related papers

• Scalable Causal Imitation Learning [48.93769483870838]
  Imitation learning enables learning a policy in an unknown environment with a latent reward signal using expert demonstrations.<n>It struggles when the imitator's and expert's observations are mismatched and unobserved confounders are present in expert demonstrations.<n>We introduce Causal Soft Q Imitation Learning (SQIL) and Causal Inverse soft-Q Learning (IQ-Learn), two off-policy causal imitation learning algorithms.
  arXiv  Detail & Related papers  (2026-07-18T23:33:11Z)
• Tailoring Strictly Proper Scoring Rules for Downstream Tasks: An Application to Causal Inference [48.78219918881965]
  We propose a principled framework for deriving task-specific strictly proper scoring rules by matching the local curvature of the downstream error metric.<n>We apply this to the Average Treatment Effect (ATE) estimation, deriving a closed-form loss and its corresponding canonical probability mapping.
  arXiv  Detail & Related papers  (2026-06-02T08:41:25Z)
• A Mechanism Study of Delayed Loss Spikes in Batch-Normalized Linear Models [28.92895415506565]
  Delayed loss spikes have been reported in neural-network training.<n>Existing theory mainly explains earlier non-monotone behavior caused by overly large fixed learning rates.<n>We study one hypothesis: normalization can postpone instability by gradually increasing the effective learning rate.
  arXiv  Detail & Related papers  (2026-04-18T03:42:05Z)
• Early Academic Capital as the Causal Origin of Dropout in Constrained Educational Systems -- Evidence from Longitudinal Data and Structural Causal Models [0.0]
  Dropout in higher education is commonly analysed through observable academic events such as course failure or repetition.<n>This study uses a causal computational social science approach to identify the origins of dropout in a constrained engineering curriculum.
  arXiv  Detail & Related papers  (2026-03-30T11:33:26Z)
• Generalized Bayes for Causal Inference [47.66564705927885]
  Uncertainty quantification is central to many applications of causal machine learning.<n>Standard Bayesian approaches typically require specifying a probabilistic model for the data-generating process.<n>We propose a generalized Bayesian framework for causal inference.
  arXiv  Detail & Related papers  (2026-03-03T14:27:23Z)
• The Causal Effect of First-Time Academic Failure on University Dropout: Evidence from a Regression Discontinuity Design [0.0]
  This study estimates the causal effect of first-time academic failure on subsequent university attrition.<n> Contrary to conventional assumptions, the results indicate that marginal first-time failure is associated with a lower probability of subsequent dropout.
  arXiv  Detail & Related papers  (2026-01-09T18:08:15Z)
• The Eminence in Shadow: Exploiting Feature Boundary Ambiguity for Robust Backdoor Attacks [51.468144272905135]
  Deep neural networks (DNNs) underpin critical applications yet remain vulnerable to backdoor attacks.<n>We provide a theoretical analysis targeting backdoor attacks, focusing on how sparse decision boundaries enable disproportionate model manipulation.<n>We propose Eminence, an explainable and robust black-box backdoor framework with provable theoretical guarantees and inherent stealth properties.
  arXiv  Detail & Related papers  (2025-12-11T08:09:07Z)
• Dual Stressors in Engineering Education: Lagged Causal Effects of Academic Staff Strikes and Inflation on Dropout within the CAPIRE Framework [0.0]
  This study provides a causal validation of the dual-stressor hypothesis in a long-cycle engineering programme in Argentina.<n>We estimate lagged causal effects of strike exposure and its interaction with inflation at entry.
  arXiv  Detail & Related papers  (2025-11-25T09:51:00Z)
• An Agent-Based Simulation of Regularity-Driven Student Attrition: How Institutional Time-to-Live Constraints Create a Dropout Trap in Higher Education [0.0]
  "The Regularity Trap" is a phenomenon where rigid assessment timelines decouple learning from accreditation.<n>We operationalize the CAPIRE framework into a calibrated Agent-Based Model (ABM) simulating 1,343 student trajectories across a 42-course Civil Engineering curriculum.<n>Results reveal that 86.4% of observed dropouts are driven by normative mechanisms (expiry cascades) rather than purely academic failure.
  arXiv  Detail & Related papers  (2025-11-20T11:21:39Z)
• The Coverage Principle: How Pre-Training Enables Post-Training [70.25788947586297]
  We study how pre-training shapes the success of the final model.<n>We uncover a mechanism that explains the power of coverage in predicting downstream performance.
  arXiv  Detail & Related papers  (2025-10-16T17:53:50Z)
• BULL-ODE: Bullwhip Learning with Neural ODEs and Universal Differential Equations under Stochastic Demand [4.285464959472458]
  We study learning of continuous-time inventory dynamics under demand and quantify when structure helps or hurts forecasting of the bullwhip effect.<n>BULL-ODE compares a fully learned Neural ODE that models the entire right-hand side against a physics-informed Universal Differential Equation (UDE)<n>Our results provide concrete guidance: enforce structure when noise is light-tailed or temporally correlated; relax structure when extreme events dominate.
  arXiv  Detail & Related papers  (2025-09-09T10:02:41Z)
• Efficient local linearity regularization to overcome catastrophic overfitting [59.463867084204566]
  Catastrophic overfitting (CO) in single-step adversarial training results in abrupt drops in the adversarial test accuracy (even down to 0%) We introduce a regularization term, called ELLE, to mitigate CO effectively and efficiently in classical AT evaluations.
  arXiv  Detail & Related papers  (2024-01-21T22:55:26Z)
• Spike No More: Stabilizing the Pre-training of Large Language Models [38.34737204307412]
  Loss spikes often occur during pre-training of large language models.<n>We explore factors to keep the gradient norm small through an analysis of the spectral norms of the Jacobian matrices for the sub-layers.<n>Our findings suggest that stabilizing the pre-training process requires two conditions: small sub-layers and large shortcut.
  arXiv  Detail & Related papers  (2023-12-28T08:53:27Z)
• Forecasting Particle Accelerator Interruptions Using Logistic LASSO Regression [62.997667081978825]
  Unforeseen particle accelerator interruptions, also known as interlocks, lead to abrupt operational changes despite being necessary safety measures. We propose a simple yet powerful binary classification model aiming to forecast such interruptions. The model is formulated as logistic regression penalized by at least absolute shrinkage and selection operator.
  arXiv  Detail & Related papers  (2023-03-15T23:11:30Z)
• Causal Expectation-Maximisation [70.45873402967297]
  We show that causal inference is NP-hard even in models characterised by polytree-shaped graphs. We introduce the causal EM algorithm to reconstruct the uncertainty about the latent variables from data about categorical manifest variables. We argue that there appears to be an unnoticed limitation to the trending idea that counterfactual bounds can often be computed without knowledge of the structural equations.
  arXiv  Detail & Related papers  (2020-11-04T10:25:13Z)
• In Search of Robust Measures of Generalization [79.75709926309703]
  We develop bounds on generalization error, optimization error, and excess risk. When evaluated empirically, most of these bounds are numerically vacuous. We argue that generalization measures should instead be evaluated within the framework of distributional robustness.
  arXiv  Detail & Related papers  (2020-10-22T17:54:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.