Search

Firms adopt randomized experiments to evaluate various interventions (e.g., website design, creative content, and pricing). However, most randomized experiments are designed to identify the impact of one specific intervention. The literature on randomized experiments lacks a holistic approach to optimize a sequence of interventions along the customer journey. Specifically, locally optimal interventions unveiled by randomized experiments might be globally suboptimal when considering their interdependence as well as the long-term rewards. Fortunately, the accumulation of a large number of historical experiments creates exogenous interventions at different stages along the customer journey and provides a new opportunity. This study integrates multiple experiments within the reinforcement learning (RL) framework to tackle the questions that cannot be answered by stand-alone randomized experiments. How can we learn optimal policy with a sequence of interventions along the customer journey based on an ensemble of historical experiments? Additionally, how can we learn from multiple historical experiments to guide future intervention trials? We propose a Bayesian recurrent Q-network model that leverages the exogenous interventions from multiple experiments to learn their effectiveness at different stages of the customer journey and optimize them for long-term rewards. Beyond optimization within the existing interventions, the Bayesian model also estimates the distribution of rewards, which can guide subject allocation in the design of future experiments to optimally balance exploration and exploitation. In summary, the proposed model creates a two-way complementarity between RL and randomized experiments, and thus, it provides a holistic approach to learning and optimizing interventions along the customer journey. This paper was accepted by Anindya Ghose, information systems. Funding: This work was supported by Adobe Faculty Research Award and the Marketing Science Institute Research Grant. Supplemental Material: The data files and online appendix are available at https://doi.org/10.1287/mnsc.2023.4914. [ABSTRACT FROM AUTHOR]

In the fields of agriculture and forestry, the Normalized Difference Vegetation Index (NDVI) is a critical indicator for assessing the physiological state of plants. Traditional imaging sensors can only collect two-dimensional vegetation distribution data, while dual-wavelength LiDAR technology offers the capability to capture vertical distribution information, which is essential for forest structure recovery and precision agriculture management. However, existing LiDAR systems face challenges in detecting echoes at two wavelengths, typically relying on multiple detectors or array sensors, leading to high costs, bulky systems, and slow detection rates. This study introduces a time-stretched method to separate two laser wavelengths in the time dimension, enabling a more cost-effective and efficient dual-spectral (600 nm and 800 nm) LiDAR system. Utilizing a supercontinuum laser and a single-pixel detector, the system incorporates specifically designed time-stretched transmission optics, enhancing the efficiency of NDVI data collection. We validated the ranging performance of the system, achieving an accuracy of approximately 3 mm by collecting data with a high sampling rate oscilloscope. Furthermore, by detecting branches, soil, and leaves in various health conditions, we evaluated the system's performance. The dual-wavelength LiDAR can detect variations in NDVI due to differences in chlorophyll concentration and water content. Additionally, we used the radar equation to analyze the actual scene, clarifying the impact of the incidence angle on reflectance and NDVI. Scanning the Red Sumach, we obtained its NDVI distribution, demonstrating its physical characteristics. In conclusion, the proposed dual-wavelength LiDAR based on the time-stretched method has proven effective in agricultural and forestry applications, offering a new technological approach for future precision agriculture and forest management. [ABSTRACT FROM AUTHOR]

Background: Lipid metabolism is a key factor in tumorigenesis and drug resistance, and models related to lipid metabolism have shown potential to predict survival and curative effects of adjuvant therapy in various cancers. However, the relationship between lipid metabolism and prognosis and treatment response of lung adenocarcinoma (LUAD) are still unclear. Methods: We enrolled seven bulk RNA-sequence datasets (GSE37745, GSE19188, GSE30219, GSE31547, GSE41271, GSE42127, and GSE72094) from the GEO database and one single-cell RNA-sequencing dataset (GSE117570) from the TISCH2 database. Non-negative matrix factorization (NMF) was utilized to construct the risk score model based on lipid score calculated by GSVA algorithm. Phs000452.v3, PMID: 26359337, PMID: 32472114, PRJEB23709 datasets were used to test the response to immunotherapy. Drug sensitivity analysis was assessed according to the GDSC database, and immunotherapy response was evaluated using the Wilcoxon test. Cellular function assays including clone formation, EDU assays and flow cytometry were implemented to explore the phenotype alteration caused by the knockdown of PTDSS1, which is one of key gene in risk score model. Results: We analyzed both bulk and single-cell RNA sequencing data to establish and validate a risk score model based on 18 lipid metabolism-related genes with significant impact on prognosis. After divided the patients into two groups according to risk score, we identified differences in lipid-related metabolic processes and a detailed portrait of the immune landscapes of high- and low-risk groups. Moreover, we investigated the potentials of our risk score in predicting response to immunotherapy and drug sensitivity. In addition, we silenced PTDSS1 in LUAD cell lines, and found that the proliferation of the cells was weakened, and the apoptosis of the cells was increased. Conclusion: Our study highlights the crucial roles of lipid metabolism in LUAD and provides a reliable risk score model, which can aid in predicting prognosis and response to immunotherapy. Furthermore, we investigated the roles of PTDSS1 in LUAD carcinogenesis, which showed that PTDSS1 regulated proliferation and apoptosis of LUAD cells. [ABSTRACT FROM AUTHOR]

Effective thermal management and electromagnetic shielding have emerged as critical goals in contemporary electronic device development. However, effectively improving the thermal conductivity and electromagnetic shielding performance of polymer composites in multiple directions continues to pose significant challenges. In this work, inspired by the efficiency of interchange bridges in enabling vehicles to pass quickly in multiple directions, we employed a straightforward method to fabricate bidirectionally oriented carbon fiber (CF)/silicone rubber composites with an interchange-bridge-like structure. The high aspect ratio of CFs and their bidirectional orientation structure play a pivotal role in facilitating the formation of thermal and electrical pathways within the composites. Meanwhile, the bidirectionally oriented CF/silicone rubber composites showed a significant enhancement in tensile strength in both the vertical and horizontal directions, attributed to the cross-arrangement of CF arrays within the composites. At a filler content of 62.3 wt%, the bidirectionally oriented CF/silicone rubber composites had a high tensile strength of 6.18 MPa. The composites also exhibited an excellent thermal conductivity of 25.3 W/(m·K) and a remarkable electromagnetic interference shielding effectiveness of 61.6 dB. The bidirectionally oriented CF/silicone rubber composites show potential for addressing thermal management and electromagnetic shielding issues in electronic devices. [ABSTRACT FROM AUTHOR]

Land use change plays a crucial role in global environmental change. Understanding the mode and land use change procedure is conducive to improving the quality of the global eco-environment and promoting the harmonized development of human–land relationships. Large river basins play an important role in areal socioeconomic development. The Yellow River Basin (YRB) is an important ecological protective screen, economic zone, and major grain producing area in China, which faces challenges with respect to ecological degradation and water and sediment management. Simulating the alterations in ecosystem service value (ESV) owing to land use change in the YRB under multiple scenarios is of great importance to guaranteeing the ecological security of the basin and improve the regional ESV. According to the land use data of 1990, 2000, 2010, and 2018, the alterations in the land use and ESV in the YRB over the past 30 years were calculated and analyzed on the basis of six land use types: cultivated land, forestland, grassland, water area, built-up land, and unused land. The patch-generating land use simulation (PLUS) model was used to simulate the land use change in the study area under three scenarios (natural development, cultivated land protection, and ecological protection in 2026); estimate the ESV under each scenario; and conduct a comparative analysis. We found that the land use area in the YRB changed significantly during the study period. The ESV of the YRB has slowly increased by ~USD 15 billion over the past 30 years. The ESV obtained under the ecological protection scenario is the highest. The simulation of the YRB's future land use change, and comparison and analysis of the ESV under different scenarios, provide guidance and a scientific basis for promoting ecological conservation and high-quality development of river basins worldwide. [ABSTRACT FROM AUTHOR]

In broad engineering fields, missing data is a common issue which often causes undesired bias and sparseness impeding rigorous data analyses. To tackle this problem, many imputation theories have been proposed and widely used. However, prior methods often require distributional assumptions and prior knowledge regarding data which may cause some difficulty for engineering research. Essentially, the fractional hot-deck imputation (FHDI) is an assumption-free imputation method, holding broad applicability in the engineering domains. FHDIs internal parameters and impact on statistical and machine learning methods, however, have been rarely understood. Thus, this study investigates the behavior and impacts of FHDI on prediction methods including generalized additive model, support vector machine, extremely randomized trees, and artificial neural network, for which four practical datasets (appliance energy, air quality, phenotypes, and weather) are used. Results show that FHDI performs better for improving the prediction accuracy compared to a simple naive method which cures missing data using the mean value of attributes, and FHDI has an asymptotically positive effect on prediction accuracy with decreasing response rates. Regarding an optimal setting, 30 to 35 is recommended for the FHDIs internal categorization number while 5 is recommended for the FHDI donors, which is aligned with Rubins recommendation. [ABSTRACT FROM AUTHOR]

On behalf of the Editorial Board, I would like to thank the following people, who acted as Reviewers during the past year. Alice Smith, Editor-in-Chief [ABSTRACT FROM AUTHOR]