@misc{yang_li_2021, title={Characterization of bonding property in asphalt pavement interlayer: A review}, volume={8}, ISSN={["2095-7564"]}, DOI={10.1016/j.jtte.2020.10.005}, abstractNote={Higher requirements are put forward for interlayer bonding property with the increase of traffic load, including bearing capacity and durability. Typical diseases caused by insufficient adhesion between layers are slippage cracking, de-bonding and pavement deformation. The correct characterization of bonding property becomes necessary with evolving concerns of interlayer failure and material innovations. In the past forty years, many researches have focused on the evaluation of interlayer bonding property, and some valuable conclusions have been drawn. In this review, the mechanism, evaluation method and influencing factors of the interlayer bond strength are reviewed. The common test equipment can be classified into the shear, pull-off and torsion test. Different influence factors are analyzed including tack coat property, temperature, asphalt aging and surface condition. It is recommended to select the appropriate tack coat rate, and apply the tack coat under the conditions of clean, dry, high surface texture and good compaction. However, the interlayer failure mechanism and the interaction between the different influencing factors need to be further studied. The future work can focus on the correlation between different test methods and evaluation parameters, which would address the lack of harmonization or consistency among various evaluation approaches. Meanwhile, it is significant to add the evaluation of interlayer bonding property into the system of pavement design.}, number={3}, journal={JOURNAL OF TRAFFIC AND TRANSPORTATION ENGINEERING-ENGLISH EDITION}, author={Yang, Kai and Li, Rui}, year={2021}, month={Jun}, pages={374–387} }
 @article{li_reich_bondell_2021, title={Deep distribution regression}, volume={159}, ISSN={["1872-7352"]}, DOI={10.1016/j.csda.2021.107203}, abstractNote={Due to their flexibility and predictive performance, machine-learning based regression methods have become an important tool for predictive modeling and forecasting. However, most methods focus on estimating the conditional mean or specific quantiles of the target quantity and do not provide the full conditional distribution, which contains uncertainty information that might be crucial for decision making. A general solution consists of transforming a conditional distribution estimation problem into a constrained multi-class classification problem, in which tools such as deep neural networks can be applied. A novel joint binary cross-entropy loss function is proposed to accomplish this goal. Its performance is compared to current state-of-the-art methods via simulation. The approach also shows improved accuracy in a probabilistic solar energy forecasting problem.}, journal={COMPUTATIONAL STATISTICS & DATA ANALYSIS}, author={Li, Rui and Reich, Brian J. and Bondell, Howard D.}, year={2021}, month={Jul} }