Weak-SIGReg: Covariance Regularization for Stable Deep Learning
arXiv:2603.05924v1 Announce Type: new Abstract: Modern neural network optimization relies heavily on architectural priorssuch as Batch Normalization and Residual connectionsto stabilize training dynamics. Without these, or in low-data regimes with aggressive augmentation, low-bias architectures like Vision Transformers (ViTs) often suffer...
Omni-Masked Gradient Descent: Memory-Efficient Optimization via Mask Traversal with Improved Convergence
arXiv:2603.05960v1 Announce Type: new Abstract: Memory-efficient optimization methods have recently gained increasing attention for scaling full-parameter training of large language models under the GPU-memory bottleneck. Existing approaches either lack clear convergence guarantees, or only achieve the standard ${\mathcal{O}}(\epsilon^{-4})$ iteration complexity...
EvoESAP: Non-Uniform Expert Pruning for Sparse MoE
arXiv:2603.06003v1 Announce Type: new Abstract: Sparse Mixture-of-Experts (SMoE) language models achieve strong capability at low per-token compute, yet deployment remains memory- and throughput-bound because the full expert pool must be stored and served. Post-training expert pruning reduces this cost, but...
Preventing Learning Stagnation in PPO by Scaling to 1 Million Parallel Environments
arXiv:2603.06009v1 Announce Type: new Abstract: Plateaus, where an agent's performance stagnates at a suboptimal level, are a common problem in deep on-policy RL. Focusing on PPO due to its widespread adoption, we show that plateaus in certain regimes arise not...
Agnostic learning in (almost) optimal time via Gaussian surface area
arXiv:2603.06027v1 Announce Type: new Abstract: The complexity of learning a concept class under Gaussian marginals in the difficult agnostic model is closely related to its $L_1$-approximability by low-degree polynomials. For any concept class with Gaussian surface area at most $\Gamma$,...
Dynamic Momentum Recalibration in Online Gradient Learning
arXiv:2603.06120v1 Announce Type: new Abstract: Stochastic Gradient Descent (SGD) and its momentum variants form the backbone of deep learning optimization, yet the underlying dynamics of their gradient behavior remain insufficiently understood. In this work, we reinterpret gradient updates through the...
DQE: A Semantic-Aware Evaluation Metric for Time Series Anomaly Detection
arXiv:2603.06131v1 Announce Type: new Abstract: Time series anomaly detection has achieved remarkable progress in recent years. However, evaluation practices have received comparatively less attention, despite their critical importance. Existing metrics exhibit several limitations: (1) bias toward point-level coverage, (2) insensitivity...
Partial Policy Gradients for RL in LLMs
arXiv:2603.06138v1 Announce Type: new Abstract: Reinforcement learning is a framework for learning to act sequentially in an unknown environment. We propose a natural approach for modeling policy structure in policy gradients. The key idea is to optimize for a subset...
Predictive Coding Graphs are a Superset of Feedforward Neural Networks
arXiv:2603.06142v1 Announce Type: new Abstract: Predictive coding graphs (PCGs) are a recently introduced generalization to predictive coding networks, a neuroscience-inspired probabilistic latent variable model. Here, we prove how PCGs define a mathematical superset of feedforward artificial neural networks (multilayer perceptrons)....
Ensemble Graph Neural Networks for Probabilistic Sea Surface Temperature Forecasting via Input Perturbations
arXiv:2603.06153v1 Announce Type: new Abstract: Accurate regional ocean forecasting requires models that are both computationally efficient and capable of representing predictive uncertainty. This work investigates ensemble learning strategies for sea surface temperature (SST) forecasting using Graph Neural Networks (GNNs), with...
Topological descriptors of foot clearance gait dynamics improve differential diagnosis of Parkinsonism
arXiv:2603.06212v1 Announce Type: new Abstract: Differential diagnosis among parkinsonian syndromes remains a clinical challenge due to overlapping motor symptoms and subtle gait abnormalities. Accurate differentiation is crucial for treatment planning and prognosis. While gait analysis is a well established approach...
FedSCS-XGB -- Federated Server-centric surrogate XGBoost for continual health monitoring
arXiv:2603.06224v1 Announce Type: new Abstract: Wearable sensors with local data processing can detect health threats early, enhance documentation, and support personalized therapy. In the context of spinal cord injury (SCI), which involves risks such as pressure injuries and blood pressure...
Gradient Flow Polarizes Softmax Outputs towards Low-Entropy Solutions
arXiv:2603.06248v1 Announce Type: new Abstract: Understanding the intricate non-convex training dynamics of softmax-based models is crucial for explaining the empirical success of transformers. In this article, we analyze the gradient flow dynamics of the value-softmax model, defined as ${L}(\mathbf{V} \sigma(\mathbf{a}))$,...
New Challenges for Federal Regulations: Executive Branch Responses
Over the last decade, federal regulations have faced increasingly more challenging hurdles. The Supreme Court’s 2024 decision in Loper Bright, putting an end to Chevron deference, and its 2022 decision in West Virginia v. EPA, announcing the “major questions doctrine,”...
Exacerbating Algorithmic Bias through Fairness Attacks
Algorithmic fairness has attracted significant attention in recent years, with many quantitative measures suggested for characterizing the fairness of different machine learning algorithms. Despite this interest, the robustness of those fairness measures with respect to an intentional adversarial attack has...
Thayerian Deference and Constitutional Interpretation
Introduction James Bradley Thayer may not be the best-known figure in the literature on constitutional interpretation, but his key ideas continue to attract attention and discussion. For over a century, scholars, judges, and Justices have been influenced by Thayer’s views...
Artificial intelligence as a medical device in radiology: ethical and regulatory issues in Europe and the United States
AI and IP: Theory to Policy and Back Again – Policy and Research Recommendations at the Intersection of Artificial Intelligence and Intellectual Property
Abstract The interaction between artificial intelligence and intellectual property rights (IPRs) is one of the key areas of development in intellectual property law. After much, albeit selective, debate, it seems to be gaining increasing practical relevance through intense AI-related market...
Georgetown Law
Georgetown Law is one of the nation's top law schools. Located in the heart of downtown Washington, D.C., it is steps away from Capitol Hill and the United ...
About the Annual Review of Criminal Procedure
Call for Submissions for Student Essay Competition
DLJ is excited to announce its second-ever online essay competition! The competition is open to all current Duke Law 2Ls who are not on an exclusive journal. Up to two winning submissions will be selected for publication in the DLJ...
Duke Law Journal’s 52nd Annual Administrative Law Symposium: Request for Proposals
The Duke Law Journal invites proposals for its 52nd Annual Administrative Law Symposium, to be held in February 2022, at Duke University School of Law in Durham, North Carolina. The Duke Law Journal’s Administrative Law Symposium has been the premier...
COVID-19 and Indian Country: A Legal Dispatch from the Navajo Nation
There has been much press coverage on the Navajo Nation’s struggle to contain the spread of COVID-19 on its lands. As of May 2, 2020, the Nation has 2,373 confirmed cases, and more than seventy deaths from the virus. These...
Protecting Our Health Care Providers from Liability in a Pandemic
While COVID-19 creates profound medical concerns for health care providers, it also creates fear of potential lawsuits. Clinicians are forced to ration scarce resources, such as ventilators, when there is an inadequate supply. Medical professionals describe chaos in hospitals that...
Closed for Business – Open for Litigation?
Can a business-closure regulation of commercial property in a pandemic be a taking? In the midst of a pandemic, it generally falls to government to enact laws and regulations in an effort to curtail the spread of disease. For example,...
When code isn’t law: rethinking regulation for artificial intelligence
Abstract This article examines the challenges of regulating artificial intelligence (AI) systems and proposes an adapted model of regulation suitable for AI's novel features. Unlike past technologies, AI systems built using techniques like deep learning cannot be directly analyzed, specified,...
Protests During the Pandemic
As a general rule, the government is permitted to restrict activities, including protesting, during the COVID-19 pandemic. The government can regulate the time, place, and manner of speech in public forums with a content neutral restriction so long as the...
Undue Computational Experimentation: Can In Silico Experiments Allows Genus Claims to Survive?
U.S. courts have, time and again, struck down genus claims for undue experimentation. The most recent blow came last year in Amgen v. Sanofi, when the Supreme Court affirmed the lower court’s ruling that Amgen’s patent on antibodies with a...