Search for: All records

Predicting the minimum operating voltage Vmin of chips stands as a crucial technique in enhancing the speed and reliability of manufacturing testing flow. However, existing Vmin prediction methods often overlook various sources of variations in both training and deployment phases. Notably, overlooking wafer zone-to-zone (intra-wafer) variations and wafer-to-wafer (inter-wafer) variations diminishes the accuracy, data efficiency, and reliability of Vmin predictors. To address this challenge, we propose Restricted Bias Alignment (RBA), a novel data-efficient Vmin prediction framework that introduces a variation alignment technique to simultaneously estimate inter- and intra-wafer variations. Furthermore, we propose utilizing class probe data to model inter-wafer variations for the first time. 

The increasing complexity of electronic systems in autonomous electric vehicles necessitates robust methods for forecasting the degradation of critical components such as printed circuit boards (PCBs). Various time series forecasting methods have been investigated to predict in-situ resistance degradation under vibration loads. However, these methods failed to capture the degradation trend under strong measurement noise. This paper introduces Monotonic Segmented Linear Regression (MSLR), a novel approach designed to capture monotonic degradation trends in time series data under significant measurement noise. By incorporating monotonic constraints, MSLR effectively models the non-decreasing behavior characteristic of degradation processes. To further enhance reliability of the prediction, we integrate Adaptive Conformal Inference (ACI) with MSLR, enabling the estimation of statistically valid upper bounds for resistance degradation with high confidence. Extensive experiments demonstrate that MSLR outperforms state-of-the-art time series forecasting baselines on real-world PCB degradation datasets. 

Accurate minimum operating voltage Vmin prediction is a critical element in manufacturing tests. Conventional methods lack coverage guarantees in interval predictions. Conformal Prediction (CP), a distribution-free machine learning approach, excels in providing rigorous coverage guarantees for interval predictions. However, standard CP predictors may fail due to a lack of knowledge of process variations. We address this challenge by providing principled conformalized interval prediction in the presence of process variations with high data efficiency, where the data from a few additional chips is utilized for calibration. We demonstrate the superiority of the proposed method on industrial 16nm chip data. 

Abstract Understanding the genetic and fitness consequences of anthropogenic bottlenecks is crucial for biodiversity conservation. However, studies of bottlenecked populations combining genomic approaches with fitness data are rare. Theory predicts that severe bottlenecks deplete genetic diversity, exacerbate inbreeding depression and decrease population viability. However, actual outcomes are complex and depend on how a species’ unique demography affects its genetic load. We used population genetic and veterinary pathology data, demographic modelling, whole-genome resequencing and forward genetic simulations to investigate the genomic and fitness consequences of a near-extinction event in the northern elephant seal. We found no evidence of inbreeding depression within the contemporary population for key fitness components, including body mass, blubber thickness and susceptibility to parasites and disease. However, we detected a genomic signature of a recent extreme bottleneck (effective population size = 6; 95% confidence interval = 5.0–7.5) that will have purged much of the genetic load, potentially leading to the lack of observed inbreeding depression in our study. Our results further suggest that deleterious genetic variation strongly impacted the post-bottleneck population dynamics of the northern elephant seal. Our study provides comprehensive empirical insights into the intricate dynamics underlying species-specific responses to anthropogenic bottlenecks. 

Predicting the minimum operating voltage Vmin of chips is one of the important techniques for improving the manufacturing testing flow, as well as ensuring the long-term reliability and safety of in-field systems. Current Vmin prediction methods often provide only point estimates, necessitating additional techniques for constructing prediction confidence intervals to cover uncertainties caused by different sources of variations. While some existing techniques offer region predictions, but they rely on certain distributional assumptions and/or provide no coverage guarantees. In response to these limitations, we propose a novel distribution-free Vmin interval estimation methodology possessing a theoretical guarantee of coverage. Our approach leverages conformalized quantile regression and on-chip monitors to generate reliable prediction intervals. We demonstrate the effectiveness of the proposed method on an industrial 5nm automotive chip dataset. Moreover, we show that the use of on-chip monitors can reduce the interval length significantly for Vmin prediction. 

Abstract Here we present 1701 light curves of 1550 unique, spectroscopically confirmed Type Ia supernovae (SNe Ia) that will be used to infer cosmological parameters as part of the Pantheon+ SN analysis and the Supernovae and H 0 for the Equation of State of dark energy distance-ladder analysis. This effort is one part of a series of works that perform an extensive review of redshifts, peculiar velocities, photometric calibration, and intrinsic-scatter models of SNe Ia. The total number of light curves, which are compiled across 18 different surveys, is a significant increase from the first Pantheon analysis (1048 SNe), particularly at low redshift ( z ). Furthermore, unlike in the Pantheon analysis, we include light curves for SNe with z < 0.01 such that SN systematic covariance can be included in a joint measurement of the Hubble constant ( H 0 ) and the dark energy equation-of-state parameter ( w ). We use the large sample to compare properties of 151 SNe Ia observed by multiple surveys and 12 pairs/triplets of “SN siblings”—SNe found in the same host galaxy. Distance measurements, application of bias corrections, and inference of cosmological parameters are discussed in the companion paper by Brout et al., and the determination of H 0 is discussed by Riess et al. These analyses will measure w with ∼3% precision and H 0 with ∼1 km s −1 Mpc −1 precision.