Profile inference revisited

Profile-guided optimization (PGO) is an important component in modern compilers. By allowing the compiler to leverage the program’s dynamic behavior, it can often generate substantially faster binaries. Sampling-based profiling is the state-of-the-art technique for collecting execution profiles in data-center environments. However, the lowered profile accuracy caused by sampling fully optimized binary often hurts the benefits of PGO; thus, an important problem is to overcome the inaccuracy in a profile after it is collected. In this paper we tackle the problem, which is also known as profile inference and profile rectification . We investigate the classical approach for profile inference, based on computing minimum-cost maximum flows in a control-flow graph, and develop an extended model capturing the desired properties of real-world profiles. Next we provide a solid theoretical foundation of the corresponding optimization problem by studying its algorithmic aspects. We then describe a new efficient algorithm for the problem along with its implementation in an open-source compiler. An extensive evaluation of the algorithm and existing profile inference techniques on a variety of applications, including Facebook production workloads and SPEC CPU benchmarks, indicates that the new method outperforms its competitors by significantly improving the accuracy of profile data and the performance of generated binaries.

Quantifying Streamflow Depletion from Groundwater Pumping: A Practical Review of Past and Emerging Approaches for Water Management

Groundwater pumping can cause reductions in streamflow (‘streamflow depletion’) that must be quantified for conjunctive management of groundwater and surface water resources. However, streamflow depletion cannot be measured directly and is challenging to estimate because pumping impacts are masked by streamflow variability due to other factors. Here, we conduct a management-focused review of analytical, numerical, and statistical models for estimating streamflow depletion and highlight promising emerging approaches. Analytical models are easy to implement, but include many assumptions about the stream and aquifer. Numerical models are widely used for streamflow depletion assessment and can represent many processes affecting streamflow, but have high data, expertise, and computational needs. Statistical approaches are a historically underutilized tool due to difficulty in attributing causality, but emerging causal inference techniques merit future research and development. We propose that streamflow depletion-related management questions can be divided into three broad categories (attribution, impacts, and mitigation) that influence which methodology is most appropriate. We then develop decision criteria for method selection based on suitability for local conditions and the management goal, actionability with current or obtainable data and resources, transparency with respect to process and uncertainties, and reproducibility.

Study and Application of Industrial Thermal Comfort Parameters by Using Bayesian Inference Techniques

This paper focuses on the use of Bayesian networks for the industrial thermal comfort issue, specifically in industries in Northern Argentina. Mined data sets that are analyzed and exploited with WEKA and ELVIRA tools are discussed. Thus, networks giving the predictive value of thermal comfort for different pairs of indoor temperature and humidity values according to activity, time, and season, verified in the workplace, were obtained. The results obtained were compared to other statistical models of linear regression used for thermal comfort, thus observing that comfort temperature values are within a same range, yet the network offered more information since a range of options for interior design parameters (temperature/relative humidity) was offered for different work, time, and season conditions. Additionally, if compared with static models of heat exchange, the contribution of Bayesian networks is noted when considering a context of actual operability and adaptability conditions to the environment, which is promising for developing thermal comfort intelligent systems, especially for the development of sustainable settings within the Industry 4.0 paradigm.

Recommendations as Treatments

In recent years, a new line of research has taken an interventional view of recommender systems, where recommendations are viewed as actions that the system takes to have a desired effect. This interventional view has led to the development of counterfactual inference techniques for evaluating and optimizing recommendation policies. This article explains how these techniques enable unbiased offline evaluation and learning despite biased data, and how they can inform considerations of fairness and equity in recommender systems.

Quantitative Analysis of the Impact of Microcredit for Women Microentrepreneurs

Microcredit has a fundamental role as a tool for financial inclusion. It is estimated that there are almost 50 million microentrepreneurs in Brazil, of which two-thirds are women. This work aims to analyze the impact of microcredit for women microentrepreneurs in comparison with men, based on data collected in northeastern Brazil in partnership Avante, a fintech specialized in microcredit. Data were collected from microentrepreneurs who had access to credit (treatment group) and those who did not have their credit granted (control group) so that they could be compared. The analysis was divided into two parts. In Part I, a descriptive statistical analysis of the collected data was performed. In Part II, some simplified causal inference techniques were applied to validate whether the impact of microcredit existed for women microentrepreneurs. The results give strong indications that women, despite having a lower income, grow more than men after access to microcredit. The annualized growth in income of women was 19.87%, while that of men was 14.66%.

Iterative deep learning for improved segmentation of endoscopic images

Iterative segmentation is a unique way to prune the segmentation maps initialized by faster inference techniques or even unsupervised traditional thresholding methods. We used our previous feedback attention-based method for this work and demonstrate that with optimal iterative procedure our method can reach competitive accuracies in endoscopic imaging. For this work, we have applied this segmentation strategy for polyps and instruments.

Lossless, scalable implicit likelihood inference for cosmological fields

We present a comparison of simulation-based inference to full, field-based analytical inference in cosmological data analysis. To do so, we explore parameter inference for two cases where the information content is calculable analytically: Gaussian random fields whose covariance depends on parameters through the power spectrum; and correlated lognormal fields with cosmological power spectra. We compare two inference techniques: i) explicit field-level inference using the known likelihood and ii) implicit likelihood inference with maximally informative summary statistics compressed via Information Maximising Neural Networks (IMNNs). We find that a) summaries obtained from convolutional neural network compression do not lose information and therefore saturate the known field information content, both for the Gaussian covariance and the lognormal cases, b) simulation-based inference using these maximally informative nonlinear summaries recovers nearly losslessly the exact posteriors of field-level inference, bypassing the need to evaluate expensive likelihoods or invert covariance matrices, and c) even for this simple example, implicit, simulation-based likelihood incurs a much smaller computational cost than inference with an explicit likelihood. This work uses a new IMNN implementation in Jax that can take advantage of fully-differentiable simulation and inference pipeline. We also demonstrate that a single retraining of the IMNN summaries effectively achieves the theoretically maximal information, enhancing the robustness to the choice of fiducial model where the IMNN is trained.

One Laptop per Child? Using Production Frontiers for Evaluating the Escuela 2.0 Program in Spain

Over the last few decades, public programs have driven the gradual adoption of information and communication technologies (ICTs) in education. The most ambitious project in Spain so far was Escuela 2.0, which provided students from the regions that opted into the program with laptops. The objective of this paper is to evaluate the impact of this program on school performance and productivity. To do this, we developed a new methodological approach based on combining causal inference techniques and the analysis of production frontiers. We calculated the differences in productivity and performance between treated and control schools using the base-group Camanho–Dyson Malmquist index and the base-group performance gap index. We estimate the impact of the program as the variation of these differences, following the essence of the difference-in-differences analysis. The main results are that Escuela 2.0 had a negative impact on performance and productivity.

The Crowd Wisdom for Location Privacy of Crowdsensing Photos

The incorporation of the mobile crowd in visual sensing provides a significant opportunity to explore and understand uncharted physical places. We investigate the gains and losses of the involvement of the crowd wisdom on users' location privacy in photo crowdsensing. For the negative effects, we design a novel crowdsensing photo location inference model, regardless of the robust location protection techniques, by jointly exploiting the visual representation, correlation, and geo-annotation capabilities extracted from the crowd. Compared with existing retrieval-based and model-based location inference techniques, our proposal poses more pernicious threats to location privacy by considering the no-reference-photos situations of crowdsensing. We conduct extensive analyses on the model with four photo datasets and crowdsourcing surveys for geo-annotation. The results indicate that being in a crowd of photos will, unfortunately, increase one's risk to be geo-identified, and highlights that the model can yield a considerable high inference accuracy (48%~70%) and serious privacy exposure (over 80% of users get privacy disclosed) with a small portion of geo-annotated samples. In view of the threats, we further propose an adaptive grouping-based signing model that hides a user's track with the camouflage of a crowd of users. Wherein, ring signature is tailored for crowdsensing to provide indistinguishable while valid identities for every user's submission. We theoretically analyze its adjustable privacy protection capability and develop a prototype to evaluate the effectiveness and performance.