Activated macrophages dominate the progression of foreign-body response (FBR) and may be in a bimodal state, which determines the fate of biomaterials postimplantation. The purpose of this study was to investigate the phenotypic profile of macrophages polarized by waterborne biodegradable polyurethane (WBPU) scaffolds with different pore diameters (PU8, PU12, and PU16) both in vitro and in vivo. The results demonstrated that WBPU scaffolds with smaller pore sizes promoted the polarization of RAW 264.7 cells towards an M1 phenotypic profile at the early stage (24 and 48 h of in vitro cultivation), indicating a pro-inflammatory response. After being implanted subcutaneously, however, the WBPU scaffolds recruited more macrophages over time and polarized them towards an M2 phenotype on Day 3 and 14, presenting an anti-inflammatory response and tissue repair. When the internal pores were filled up (on Day 30 of implantation), the interaction between the scaffolds and macrophages decreased, indicating an endpoint of tissue repair. In general, WBPU scaffolds with tunable internal pore sizes have potential application prospects in the field of tissue engineering.

In view of the common and difficult to solve vehicle vibration problem, taking a large truck manufacturing enterprise in China as an example, an improved model for solving truck vibration problem is established by using the G8D method, and the unbalanced excitation force of the wheel system is analysed. The coupling of the excitation frequency and the natural frequency of the system leads to the resonance phenomenon, as well as the inadequate damping function of the system, is the fundamental cause of the vibration problem. After verifying and implementing permanent correction measures at the same time at the three levels of components, devices, and the entire vehicle, the acceleration of the seat guide rails for the vibration performance of the truck reduces from the original state 1.04 m/s2 to 0.6 m/s2, a decrease of 42.3%, which reaches the best level of mainstream cars in the country and is close to the optimal level of 0.5 m/s2 among the same kinds of cars in Germany. Therefore, the improved model can improve the sustainability of product manufacturing, provide industry guidance for solving the quality problem of truck vibration, and provide a sustainable guarantee for social public transport safety.

The dynamic job-shop scheduling problem (DJSP) is a class of scheduling tasks that specifically consider the inherent uncertainties such as changing order requirements and possible machine breakdown in realistic smart manufacturing settings. Since traditional methods cannot dynamically generate effective scheduling strategies in face of the disturbance of environments, we formulate the DJSP as a Markov decision process (MDP) to be tackled by reinforcement learning (RL). For this purpose, we propose a flexible hybrid framework that takes disjunctive graphs as states and a set of general dispatching rules as the action space with minimum prior domain knowledge. The attention mechanism is used as the graph representation learning (GRL) module for the feature extraction of states, and the double dueling deep Q-network with prioritized replay and noisy networks (D3QPN) is employed to map each state to the most appropriate dispatching rule. Furthermore, we present Gymjsp, a public benchmark based on the well-known OR-Library, to provide a standardized off-theshelf facility for RL and DJSP research communities. Comprehensive experiments on various DJSP instances confirm that our proposed framework is superior to baseline algorithms with smaller makespan across all instances and provide empirical justification for the validity of the various components in the hybrid framework.

<p>Research in artificial intelligence demonstrates the applicability and flexibility of the reinforcement learning (RL) technique for the dynamic job shop scheduling problem (DJSP). However, the RL-based method will always overfit to the training environment and cannot generalize well to novel unseen situations at deployment time, which is unacceptable in real-world production. For this reason, this paper proposes a highly generalizable reinforcement learning framework named Train Once For All (TOFA) for the dynamic job shop scheduling problem. The trivial and non-trivial states are distinguished when the DJSP is formulated as a semi-Markov decision process, defining the size-agnostic state, action, and reward function. A novel graph representation learning method based on attention mechanism and spatial pyramid pooling is implemented to compress the disjunctive graphs of differentsize DJSP into fixed-length feature vectors. Combining the proposed dynamic frame skipping and an improved prioritized experience replay method that considers the sample quality difference at different training phases. TOFA shows superb generalization capability, outperforms practically favored dispatching rules and even instance-by-instance training RL-based schedulers on various benchmark DJSP. Additionally, we proved that TOFA acquires a transferable scheduling policy that can be used to schedule a whole new DJSP without additional training.</p>

No abstract

An impulsive Lotka-Volterra type predator-prey model with prey dispersal in two-patch environments and time delays is investigated, where we assume the model of patches with a barrier only as far as the prey population is concerned, whereas the predator population has no barriers between patches. By applying the continuation theorem of coincidence degree theory and by means of a suitable Lyapunov functional, a set of easily verifiable sufficient conditions are obtained to guarantee the existence, uniqueness, and global stability of positive periodic solutions of the system. Some known results subject to the underlying systems without impulses are improved and generalized. As an application, we also give two examples to illustrate the feasibility of our main results.

<p>Research in artificial intelligence demonstrates the applicability and flexibility of the reinforcement learning (RL) technique for the dynamic job shop scheduling problem (DJSP). However, the RL-based method will always overfit to the training environment and cannot generalize well to novel unseen situations at deployment time, which is unacceptable in real-world production. For this reason, this paper proposes a highly generalizable reinforcement learning framework named Train Once For All (TOFA) for the dynamic job shop scheduling problem. The trivial and non-trivial states are distinguished when the DJSP is formulated as a semi-Markov decision process, defining the size-agnostic state, action, and reward function. A novel graph representation learning method based on attention mechanism and spatial pyramid pooling is implemented to compress the disjunctive graphs of differentsize DJSP into fixed-length feature vectors. Combining the proposed dynamic frame skipping and an improved prioritized experience replay method that considers the sample quality difference at different training phases. TOFA shows superb generalization capability, outperforms practically favored dispatching rules and even instance-by-instance training RL-based schedulers on various benchmark DJSP. Additionally, we proved that TOFA acquires a transferable scheduling policy that can be used to schedule a whole new DJSP without additional training.</p>

When the dimension of covariates in the regression model is high, one usually uses a submodel as a working model that contains significant variables. But it may be highly biased and the resulting estimator of the parameter of interest may be very poor when the coefficients of removed variables are not exactly zero. In this paper, based on the selected submodel, we introduce a two-stage remodeling method to get the consistent estimator for the parameter of interest. More precisely, in the first stage, by a multistep adjustment, we reconstruct an unbiased model based on the correlation information between the covariates; in the second stage, we further reduce the adjusted model by a semiparametric variable selection method and get a new estimator of the parameter of interest simultaneously. Its convergence rate and asymptotic normality are also obtained. The simulation results further illustrate that the new estimator outperforms those obtained by the submodel and the full model in the sense of mean square errors of point estimation and mean square prediction errors of model prediction.