2578 papers
This paper analyzes how endpoint definitions and censoring strategies significantly impact battery lifetime prediction accuracy across different data sources, advocating for a standardized engineering evaluation procedure that explicitly declares references, separates observed and censored outcomes, and reports cohort-level error bounds to ensure reliable comparisons.
This paper presents a memristive spatiotemporal compression architecture that achieves efficient visual motion prediction by integrating temporal changes into conductance states and compressing spatial data into one-dimensional currents, resulting in significantly reduced latency and energy consumption compared to GPU baselines while maintaining high accuracy.
This paper proposes that rather than viewing artificial intelligence as a disruption to standardized patient (SP) methodology, it should be embraced as a complementary modality that, when integrated with human SPs under professional oversight, expands scalable, personalized learning opportunities while preserving the core educational principles and human-centered values of clinical skills education.
This study quantitatively determines the takeoff throttle threshold of an ArduPilot Gazebo quadrotor simulation to be between commands 1485 and 1486 for a 1.98 kg model, establishing a baseline for future physical implementation and calibration.
This study validates a numerical model for electro-osmotic drainage in weakly expansive soil slopes using perforated tubular electrodes through laboratory proof-of-concept tests and a globally sampled uncertainty analysis, revealing that cathodic hydraulic boundary conditions are the primary design uncertainty while intermittent operation can improve energy efficiency.
This study experimentally establishes a weldability transition map for DCEN-GTAW butt joining of 0.08 mm AA3003 aluminum foil, demonstrating that stable welding windows are critically dependent on the interplay between welding current and root gap, with optimal parameters identified as 5–10 A for a 1.0 mm gap and 5–9 A for a 2.0 mm gap.
This study utilizes finite element analysis to demonstrate that externally confining rectangular RCC–timber composite columns with a 2 mm mild steel tube significantly enhances their axial load capacity, stiffness, and ductility while mitigating brittle failure modes compared to unconfined counterparts.
This study demonstrates that machine-learning postprocessing of HEC-RAS 2D flood simulations in southeastern Louisiana significantly reduces water-level errors, with the inclusion of current water-surface elevation providing the most consistent improvement and recent historical data offering additional, though less uniform, benefits.
This study presents an interpretable hybrid surrogate model that combines symbolic regression with Gaussian Process residual correction to accurately predict diameter deviations in injection molding under limited experimental data, achieving improved accuracy and providing a transparent framework for process optimization.
This paper proposes a model-independent pre-training framework using Wasserstein-based distributional distances to quantify source-target shifts in petroleum engineering data, aiming to predict zero-shot model transferability, though initial synthetic experiments yielded inconclusive correlations between the shift index and performance degradation.