Lower Estimates Research Articles

MVImgNet2.0: A Larger-scale Dataset of Multi-view Images

MVImgNet is a large-scale dataset that contains multi-view images of ~220k real-world objects in 238 classes. As a counterpart of ImageNet, it introduces 3D visual signals via multi-view shooting, making a soft bridge between 2D and 3D vision. This paper constructs the MVImgNet2.0 dataset that expands MVImgNet into a total of ~520k objects and 515 categories, which derives a 3D dataset with a larger scale that is more comparable to ones in the 2D domain. In addition to the expanded dataset scale and category range, MVImgNet2.0 is of a higher quality than MVImgNet owing to four new features: (i) most shoots capture 360° views of the objects, which can support the learning of object reconstruction with completeness; (ii) the segmentation manner is advanced to produce foreground object masks of higher accuracy; (iii) a more powerful structure-from-motion method is adopted to derive the camera pose for each frame of a lower estimation error; (iv) higher-quality dense point clouds are reconstructed via advanced methods for objects captured in 360 ° views, which can serve for downstream applications. Extensive experiments confirm the value of the proposed MVImgNet2.0 in boosting the performance of large 3D reconstruction models. MVImgNet2.0 will be public at luyues.github.io/mvimgnet2 , including multi-view images of all 520k objects, the reconstructed high-quality point clouds, and data annotation codes, hoping to inspire the broader vision community.

Genetic Variation in Yield and Fruit Weight Among Strawberry (Fragaria ×ananassa) Cultivars and the Interaction With Year and Location Effect

ABSTRACTStrawberry (Fragaria ×ananassa) is an important crop in the United States. In this study, we analysed strawberry trials established across 7 years and two locations. Our objective was to estimate genetic parameters and predict breeding values for yield. Linear mixed models were used for unbalanced data across years and locations and study the genotype by year and location interactions. The results suggested significant differences between cultivars for fruit weight but not for total and marketable yield. Fruit weight was controlled by genetic factors, shown by the high heritability of cultivar means (0.93), whereas marketable yield and total yield showed lower heritability estimates (0.48 and 0.63, respectively), indicating a substantial environmental influence. For fruit weight, cultivars exhibited stability across years and locations, supported by the high additive genetic correlations observed for cultivar‐by‐year (r = 0.90) and cultivar‐by‐location interactions (r = 0.98). Fruit weight and total yield were genetically independent as suggested near zero additive genetic correlation (−0.06 ± 0.416) between two traits, and they should be considered for an index selection in the strawberry breeding strategy.

Evaluation of traditional and bootstrapped methods for assessing data-poor fisheries: a case study on tropical seabob shrimp (Xiphopenaeus kroyeri) with an improved length-based mortality estimation method.

Unrealistic model assumptions or improper quantitative methods reduce the reliability of data-limited fisheries assessments. Here, we evaluate how traditional length-based methods perform in estimating growth and mortality parameters in comparison with unconstrained bootstrapped methods, based on a virtual population and a case study of seabob shrimp (Xiphopenaeus kroyeri, Heller, 1862). Size data were obtained for 5,725 seabob shrimp caught in four distinct fishing grounds in the Southwestern Atlantic. Also, a synthetic population with known parameter values was simulated. These datasets were analyzed using different length-based methods: the traditional Powell-Wetheral plot method and novel bootstrapped methods. Analysis with bootstrapped ELEFAN (fishboot package) resulted in considerably lower estimates for asymptotic size (L ∞), instantaneous growth rate (K), total mortalities (Z) and Z/K values compared to traditional methods. These parameters were highly influenced by L ∞ estimates, which exhibited median values far below maximum lengths for all samples. Contrastingly, traditional methods (PW method and L max approach) resulted in much larger L ∞ estimates, with average bias >70%. This caused multiplicative errors when estimating both Z and Z/K, with an astonishing average bias of roughly 200%, with deleterious consequences for stock assessment and management. We also present an improved version of the length-converted catch-curve method (the iLCCC) that allows for populations with L ∞ > L max and propagates the uncertainty in growth parameters into mortality estimates. Our results highlight the importance of unbiased growth estimates to robustly evaluate mortality rates, with significant implications for length-based assessments of data-poor stocks. Thus, we underscore the call for standardized, unconstrained use of fishboot routines.

New estimates for Hermite–Hadamard–Fejer-type inequalities containing Raina fractional integrals

The Hermite–Hadamard–Fejér-type inequality is an effective utensil for examining upper and lower estimations of the integrals of convex functions. In this study, the power mean inequality and Hölder inequality are employed. The outcomes are new Hermite–Hadamard–Fejér-type inequalities via Raina fractional integrals for m-convex functions. The findings are extensions of many previous investigations.

A novel hypoxia- and lactate metabolism-related prognostic signature to characterize the immune landscape and predict immunotherapy response in osteosarcoma.

Immunotherapy has shown considerable promise in cancer treatment, yet only a minority of osteosarcoma patients derive benefits from this approach. Hypoxia and lactate metabolism are two predominant characteristics of the tumor microenvironment. These features are crucial for molding the immune landscape and thus have the potential to act as predictive indicators for immunotherapy response. Prognostic modeled genes were identified through univariate and multivariate Cox regression as well as LASSO regression analyses. The tumor microenvironment was evaluated using ESTIMATE, CIBERSORT, and ImmuCellAI analyses. Tide prediction and expression of immune checkpoints, MHC molecules, chemokines, interleukins, interferons, receptors, and other cytokines were utilized to estimate immunotherapy efficacy. Single-cell analysis was performed to demonstrate the expression of modeled genes among various immune cell types. Experimental validation was carried out to verify the expression and functions of SFXN4 and SQOR. A potent signature was constructed with 8 genes related to hypoxia and lactate metabolism, including MAFF, COL5A2, FAM162A, SQOR, UQCRB, SFXN4, PFKFB2 and COX6A2. A nomogram incorporating risk scores and other clinical features demonstrated excellent predictive capacity. Osteosarcoma patients with high-risk scores exhibited poor prognosis and more "cold" tumor characteristics. According to the ESTIMATE algorithm, these patients displayed lower immune, stromal, and ESTIMATE scores, partially attributed to inadequate infiltration of key immunocytes. The Ciborsort analysis similarly indicated that high-risk individuals had diminished infiltration of critical anti-tumor immune cells such as Cytotoxic T cells, CD4+ T cells, and NK cells. The low expression levels of certain immune checkpoints, MHC molecules, chemokines, interleukins, interferons, receptors, and other cytokines in high-risk cases suggested their unsatisfactory responses to immune treatment. Tide prediction further demonstrated that fewer individuals classified as high risk may exhibit sensitivity to immune checkpoint inhibitor therapy. Notably, SFXN4 was found to be highly expressed in osteosarcoma tissues and cells; it promoted the growth, migration, and invasion of osteosarcoma cells, while SQOR had the opposite effect. Our research has developed a robust hypoxia- and lactate metabolism-related gene signature, providing a solid theoretical foundation for prognosis prediction, classification of "cold" and "hot" tumors, accessing immunotherapy response, and directing personalized treatment for osteosarcoma.

A Dual-Stream Fusion Network for Human Energy Expenditure Estimation with Wearable Sensor

With the increasing awareness of health, using wearable sensors to monitor individual activities and accurately estimate energy expenditure has become a current research focus. However, existing research encounters challenges including low estimation accuracy, a deficiency of frequency domain features, and difficulty in integrating time domain and frequency domain features. To address these issues, we propose an innovative framework called the Dual-Stream Fusion Network (DSFN). This framework combines the Time Domain Encoding (TDE) module, the Frequency Domain Hierarchical-Split Encoding (FDHSE) module, and a Two-Stage Feature Fusion (TSF) module. Specifically, the temporal stream of the framework employs the TDE module to capture deep temporal features that reflect the complex dynamic variations in time-series data. The frequency domain stream introduces the FDHSE module, which extracts frequency domain features using a multi-level, multi-scale approach, ensuring a comprehensive and diverse representation of frequency information. Through this dual-stream architecture, our model effectively learns both time and frequency domain features, addressing the limitations of frequency domain features observed in prior studies. Additionally, we propose the TSF module to fully integrate time and frequency domain features, effectively overcoming the challenge of fusing these two types of features. We conducted experiments on two public datasets, namely the GOTOV dataset (elderly people) and the JSI dataset (young people). Experimental results demonstrate that our method achieves excellent performance across different age groups. Compared to the baseline models, the proposed DSFN significantly improves the accuracy of human energy expenditure estimation.

How user language affects conflict fatality estimates in ChatGPT

OpenAI’s ChatGPT language model has gained popularity as a powerful tool for problem-solving and information retrieval. However, concerns arise about the reproduction of biases present in the language-specific training data. In this study, we address this issue in the context of the Israeli–Palestinian and Turkish–Kurdish conflicts. Using GPT-3.5, we employed an automated query procedure to inquire about casualties in specific airstrikes, in both Hebrew and Arabic for the former conflict and Turkish and Kurdish for the latter. Our analysis reveals that GPT-3.5 provides 34 ± 11% lower fatality estimates when queried in the language of the attacker than in the language of the targeted group. Evasive answers denying the existence of such attacks further increase the discrepancy. A simplified analysis on the current GPT-4 model shows the same trends. To explain the origin of the bias, we conducted a systematic media content analysis of Arabic news sources. The media analysis suggests that the large-language model fails to link specific attacks to the corresponding fatality numbers reported in the Arabic news. Due to its reliance on co-occurring words, the large-language model may provide death tolls from different attacks with greater news impact or cumulative death counts that are prevalent in the training data. Given that large-language models may shape information dissemination in the future, the language bias identified in our study has the potential to amplify existing biases along linguistic dyads and contribute to information bubbles.

Performance Evaluation of QT-RR Adaptation Time Lag Estimation in Exercise Stress Testing.

Slower adaptation of the QT interval to sudden changes in heart rate has been identified as a risk marker of ventricular arrhythmia. The gradual changes observed in exercise stress testing facilitates the estimation of the QT-RR adaptation time lag. The time lag estimation is based on the delay between the observed QT intervals and the QT intervals derived from the observed RR intervals using a memoryless transformation. Assuming that the two types of QT interval are corrupted with either Gaussian or Laplacian noise, the respective maximum likelihood time lag estimators are derived. Estimation performance is evaluated using an ECG simulator which models change in RR and QT intervals with a known time lag, muscle noise level, respiratory rate, and more. The accuracy of T-wave end delineation and the influence of the learning window positioning for model parameter estimation are also investigated. Using simulated datasets, the results show that the proposed approach to estimation can be applied to any changes in heart rate trend as long as the frequency content of the trend is below a certain frequency. Moreover, using a proper position of the learning window for exercise so that data compensation reduces the effect of nonstationarity, a lower mean estimation error results for a wide range of time lags. Using a clinical dataset, the Laplacian-based estimator shows a better discrimination between patients grouped according to the risk of suffering from coronary artery disease. Using simulated ECGs, the performance evaluation of the proposed method shows that the estimated time lag agrees well with the true time lag.

Data-driven parallel optimization method for sensor placement and state estimation based on generative adversarial principle in unobservable systems

Abstract Data-Driven State Estimation (DDSE) is characterized by low measurement requirements and high estimation accuracy, which provides the possibility of stable and efficient estimation for unobservable distribution systems with low measurement quantities. However, the lack of an efficient and flexible sensor placement method with DDSE at this stage hinders the development and application of DDSE in distribution networks. To address this issue, a data-driven parallel optimization method for sensor placement and state estimation based on the generative adversarial principle in unobservable systems is proposed in this paper. This method transforms sensor placement and state estimation into an adversarial learning model, solvable through gradient descent. The sensor placement optimization strategy generates a cost-effective placement scheme, while DDSE evaluates the sensor scheme, provides improvement suggestions, and enhances its estimation accuracy. Through adversarial training, the proposed scheme yields a significantly reduced number of sensor placement schemes compared to traditional methods, while maintaining high precision of DDSE. Simulation results on the IEEE-33 node system validate the effectiveness of the proposed approach.

Sea surface heat flux helps predicting thermocline in the South China Sea

In this study, a deep learning model called Four Dimensional Residual Network (4D-ResNet) was proposed, which can capture both temporal and spatial information. Temperatures at various depths were predicted for the next 40 days using the last month's sea surface variables, and a spatio-temporal prediction of the thermocline was achieved. In addition to the satellite-observed sea surface parameters: sea surface temperature (SST), sea level anomaly (SLA), and sea surface wind (SSW), net heat flux (Qnet) was also included in the model input. Qnet can alter the density of the upper water, resulting in convection or improved stratification stability. The results indicate that the additional input of Qnet improves the model's accuracy, especially at the depth of the thermocline, where the RMSE reduced by up to 13.7%. The 4D-ResNet model has much lower estimation error compared to other models and successfully captures the seasonal characteristics of the thermocline.

Lithium battery SOC estimation based on improved sparrow search algorithm and backpropagation neural network

Accurate state-of-charge (SOC) estimation is crucial for optimal battery management. This paper proposes a novel method, the Improved Sparrow Search Algorithm-Backpropagation (ISSA-BP) neural network, to address the issue of low estimation accuracy encountered with a single BP neural network. ISSA is used to optimize the initial weights and thresholds of the BP neural network, effectively overcoming its tendency to get stuck in local minima. Compared to the single BP neural network, ISSA-BP demonstrates significantly improved accuracy under two conditions (DST and BJDST), with reductions in root mean square error by 64.0% and 50.9% and mean absolute error by 69.8% and 51.1%, respectively. These results highlight the superior robustness and accuracy of the ISSA-BP algorithm for SOC estimation in lithium batteries.

Prediction of cardiovascular events and all-cause mortality using race and race-free estimated glomerular filtration rate in African Americans: the Jackson Heart Study.

New Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations without a race adjustment were developed to estimate the glomerular filtration rate (eGFR). We aimed to compare the performance of five CKD-EPI eGFR equations, with or without race, in predicting cardiovascular disease (CVD) events and all-cause mortality in Black Americans from the Jackson Heart Study. JHS is an ongoing population-based prospective cohort study of African Americans in the Jackson, Mississippi, metropolitan area. Five CKD-EPI equations were used to estimate GFR at baseline using serum creatinine (Cr) or cystatin C (cys), including 2009 eGFRcr(ASR [age, sex, race]), 2021 eGFRcr(AS [age and sex]), 2012 eGFRcr-cys(ASR), 2021 eGFRcr-cys(AS), 2012 eGFRcys(AS). Endpoints were incident CVD events and all-cause mortality. Cox proportional hazards regression was used to assess the associations between different eGFRs and outcomes adjusting for atherosclerotic risk factors. Harrell's C-statistics and Net Reclassification Index (NRI) were used to assess the predictive utility. Among 5,129 participants (average age 54.8 ± 12.8 yrs), 1898 were male (37.0%). eGFRcr(AS) provided lower estimates and resulting in a greater proportion of participants categorized as CKD than eGFRcr(ASR), eGFRcr-cys(ASR), eGFRcr-cys(AS) and eGFRcys(AS). A median follow-up of 13.7 and 14.3 years revealed 411 (9.3%) CVD incidents and 1,207 (23.5%) deaths. Lower eGFRs were associated with CVD incidents and all-cause mortality. eGFRcr-cys(ASR), eGFRcr-cys(AS) and eGFRcys(AS) were strongly associated with incident CVD events and all-cause mortality than eGFRcr(ASR) and eGFRcr(AS). A significant discrimination improvement was found in C-statistics for predicting incident CVD events and all-cause mortality after adding each eGFR measure to the basic model including atherosclerotic risk factors. Across a 7.5% 10-year risk threshold, eGFRcys(AS) improved net classification of all-cause mortality (NRI: 2.19, 95%CI: 0.08, 4.65%). eGFR based on creatinine omit race has the lowest mean and detects more CKD patients in Black population. The eGFRs incorporating cystatin C strengthens the association between the eGFR and the risks of incident CVD and all-cause mortality. Cystatin C-based eGFR equations might be more appropriate for predicting CVD and mortality among Black population.

Eкспериментально-розрахункове визначення коефіцієнтів гідродинамічної моделі кавітуючих насосів рідинних ракетних двигунів

Cavitation phenomena in liquid-propellant rocket engine (LPRE) pumps not only affect the power performance characteristics of the pumps, but they also affect the pump dynamics and pogo vibrations. The theoretical characterization of cavitation phenomena in LPRE pumps is not a widely used practice because theoretical and experimental data are in unsatisfactory agreement. Because of this, use is made of approaches that employ experimental data. The goal of this work is to determine the coefficients of a hydrodynamic model of cavitating LPRE pumps throughout the cavity existence region based on the experimental frequencies of cavitation oscillations and cavitation self-oscillation boundaries. In determining the cavity elasticity and negative resistance, use was made of the experimental cavitation oscillation frequencies of 26 LPRE pumps differing in dimensions and capacity. In determining the cavitation resistance distribution coefficient and the cavity-due disturbance transfer time, the experimental cavitation self-oscillation boundaries of 14 more pumps were used. To extend the cavity elasticity determination region, the extrapolation dependence of the cavity elasticity in cavitation stall regimes was updated. To make the stratification of the cavity resistance dependence more uniform in the range of large discharge coefficients, incipient cavitation numbers were refined. Using he qualitative dependence of the cavitation resistance distribution coefficient obtained from theoretical transfer matrices of cavitating pumps and its lower estimate (at zero disturbance transfer time) and upper estimate (for a uniform stratification of pump transfer matrix determinants), its analytical dependence was found. Using it and the coefficients of a mathematical model of cavitation oscillations on the cavitation-self oscillation boundary, disturbance transfer times were found and approximated.

Estimation of function's supports under arithmetic constraints

AbstractThe well-known inequality $$\lvert {\rm supp}(f) \rvert \lvert {\rm supp}( \widehat f) \rvert \geq |G|$$ | supp ( f ) | | supp ( f ^ ) | ≥ | G | gives a lower estimation for each support. In this paper we consider the case where there exists a slowly increasing function $$F$$ F such that $$\lvert {\rm supp}(f) \rvert \leq F(\lvert {\rm supp}( \widehat f) \rvert )$$ | supp ( f ) | ≤ F ( | supp ( f ^ ) | ) . We will show that this can be done under some arithmetic constraint. The two links that help us come from additive combinatorics and theoretical computer science. The first is the additive energy which plays a central role in additive combinatorics. The second is the influence of Boolean functions. Our main tool is the spectral analysis of Boolean functions. We prove an uncertainty inequality in which the influence of a function and its Fourier spectrum play a role.

Multiparametric risk predictions for morbidity and mortality in patients with pulmonary arterial hypertension: insights from a nation-wide registry

Abstract Background and aims In this retrospective analysis of a nation-wide registry on pulmonary arterial hypertension (PAH), group IV and V, we aimed to evaluate risk predictions for morbidity and mortality. Methods This study included 903 patients enrolled from 24 cardiology centers participating in the RegiStry on clInical outcoMe and sUrvival in pulmonaRy hypertension Groups (SIMURG II). ESC/ERS 2022 risk scoring, COMPERA 2.0 and REVEAL lite 2 models were used for baseline risk predictions, and COMPERA 2.0 model was used for risk evaluation at follow-up. Composite endpoint (CEP) definitions were adopted from the SERAPHIN and GRIPHON trials. Results Age (mean +SD) was 51.27 + 21.37 years, and 72 % of patients were female. Incident cases were noted in 65.9 % of overall study group. Idiopathic PAH (IPAH), PAH-associated with congenital heart diseases (CHD-PAH), PAH-associated with connective tissue diseases (CTD-PAH), porto-PAH, Group IV and V pulmonary hypertension (PH) were documented in 33 %, 33.5 %, 12 %, 1.6 %, 20 % and 1.7 % of overall patients, respectively. Background mono and dual targeted combinations, and sequential dual or triple combinations were noted in 17 %, 57 % and 26 % of patients, respectively. According to the ESC/ERS 2022 baseline risk scoring, low-, intermediate-, and high-risk were noted in 1.2 %, 14.1 %, and 84.7 %, respectively. Median follow-up time was 867 (412-1667) days. Overall rates of mortality and CEP were 26 % and 39%, respectively. Prevalent versus incident cases were associated signifinatly lower mortality and CEP ( p=0.028 and p=0.024, respectively). Mortality and CEP rates were 31 % and 44 % in IPAH, 20 % and 39 % in CHD-PAH, 31 % and 43 % in CTD-PAH, and 25 % and 30 % in group IV patients, respectively. Baseline high versus intermediate /low risk status were associated with a lower survival estimates for mortality and CEP in patients with PAH ( p=0.03 and p=0.003, respectively). Baseline COMPERA 2.0 risk status revealed significant differences in overall survival and CEP-free survival among risk groups ( p<0.0001 for both), but 1st year COMPERA 2.0 score did not (p=0.39 and p=0.06, respectively) . As compared to baseline, 1st year COMPERA 2.0 score decreased in 31 %, and remained unchanged in 34 % of patients. Baseline REVEAL lite 2.0 score was associated with mortality (hazard ratio (HR): 1.14;1.08-1.2, p<0.0001) and CEP (HR:1.17;1.1-1.25, p<0.0001) risks. Male sex (HR: 1.73; 1.2-2.51, p = 0.004), six-minute walk distance (p<0.001), functional class III/IV (HR: 2.52;1.47-4.32, p<0.001) and baseline low-risk status (HR:0.37;0.21-0.65, p<0.001) were independent predictors of mortality. Conclusions Our nation-wide data revealed current insights concerning the multiparametric risk predictions for morbidity and mortality in PAH.

Filtered Lie-Trotter splitting for the “good” Boussinesq equation: Low regularity error estimates

We investigate a filtered Lie-Trotter splitting scheme for the “good” Boussinesq equation and derive an error estimate for initial data with very low regularity. Through the use of discrete Bourgain spaces, our analysis extends to initial data in H s H^{s} for 0 > s ≤ 2 0>s\leq 2 , overcoming the constraint of s > 1 / 2 s>1/2 imposed by the bilinear estimate in smooth Sobolev spaces. We establish convergence rates of order τ s / 2 \tau ^{s/2} in L 2 L^2 for such levels of regularity. Our analytical findings are supported by numerical experiments.

Upper and lower estimates for rate of convergence in the Chernoff product formula for semigroups of operators

Upper and lower estimates for rate of convergence in the Chernoff product formula for semigroups of operators

Estimation of genetic parameters for reproductive traits in Korean dairy cattle.

In Korea, dairy cattle breeding programs have historically prioritized productive, conformation traits, leading to positive improvements, yet reproductive traits have lagged in development. This study was conducted to develop the breeding program of key reproductive traits in the Korean dairy cattle population. Utilizing data from 7,596 farms and over seven million observations, we conducted quality control to rectify manual entry errors and selected traits in line with international genetic evaluation standards. Traits analyzed included heifer conception rate (HCR), interval from calving to first insemination (CF), cow conception rate (CCR), interval from first to last insemination (FL), and days open (DO). Genetic parameters were estimated using a single trait animal model for HCR and a multiple lactation animal model for CF, CCR, FL, and DO, considering contemporary group of herd-insemination year, insemination month, and monthly age (MA) as fixed effects. Results showed low heritability estimates, ranging from 0.007 to 0.035 across different traits and lactations. Theoretical reliability appears to be low on average due to the influence of heritability, but it showed sufficiently high reliability in some sires (over 0.8). In terms of genetic and phenotypic trends, capacity for reproductive traits declined for a long time until around 2014. In recent individuals, improved trend can be found. This study addressed the critical need for enhancing reproductive efficiency to complement the existing breeding goals, thereby supporting sustained economic viability in the dairy industry. The results underscore the need for improved data quality and methodological adjustments for reproduction records to enhance the genetic evaluation of dairy cattle in Korea.

Socioeconomic status, smoking, and lung cancer: mediation and bias analysis in the SYNERGY study.

Increased lung-cancer risks for low socioeconomic status (SES) groups are only partially attributable to smoking habits. Little effort has been made to investigate the persistent risks related to low SES by quantification of potential biases. Based on 12 case-control studies, including 18 centers of the international SYNERGY project (16,550 cases, 20,147 controls), we estimated controlled direct effects (CDE) of SES on lung cancer via multiple logistic regression, adjusted for age, study center, and smoking habits, and stratified by sex. We conducted mediation analysis by inverse odds ratio weighting to estimate natural direct effects (NDE) and natural indirect effects via smoking habits. We considered misclassification of smoking status, selection bias, and unmeasured mediator-outcome confounding by genetic risk, both separately as well as by multiple quantitative bias analysis, using bootstrap to create 95% simulation intervals (SI). Mediation analysis of lung-cancer risks for SES estimated mean proportions of 43% in men and 33% in women attributable to smoking. Bias analyses decreased direct effects of SES on lung cancer, with selection bias showing the strongest reduction in lung-cancer risk in the multiple bias analysis. Lung-cancer risks remained increased for lower SES groups, with higher risks in men [4th versus 1st (highest) SES quartile: CDE 1.50 (SI 1.32-1.69)] than women [CDE 1.20 (SI 1.01-1.45)]. NDE were similar to CDE, particularly in men. Bias adjustment lowered direct lung-cancer risk estimates of lower SES groups. However, risks for low SES remained elevated, likely attributable to occupational hazards or other environmental exposures.

Understanding Power Gating Mechanism Based on Workload Classification of Modern Heterogeneous Many-Core Mobile Platform in the Dark Silicon Era

The rapid progress in mobile computing necessitates energy efficient solutions to support substantially diverse and complex workloads. Heterogeneous many core platforms are progressively being adopted in contemporary embedded implementations for high performance at low power cost estimations. These implementations experience diverse workloads that offer drastic opportunities to improve energy efficiency. In this paper, we propose a novel per core power gating (PCPG) approach based on workload classifications (WLC) for drastic energy cost minimization in the dark silicon era. Core of our paradigm is to use an integrated sleep mode management based on workloads classification indicated by the performance counters. A number of real applications benchmark (PARSEC) are adopted as a practical example of diverse workloads, including memory- and CPU-intensive ones. In this paper, these applications are exercised on Samsung Exynos 5422 heterogeneous many core system showing up to 37% to 110% energy efficient when compared with our most recent published work, and ondemand governor, respectively. Furthermore, we illustrate low-complexity and low-cost runtime per core power gating algorithm that consistently maximize IPS/Watt at all state space.