Conservative-Progressive Collaborative Learning for Semi-Supervised Semantic Segmentation.

Pseudo supervision is regarded as the core idea in semi-supervised learning for semantic segmentation, and there is always a tradeoff between utilizing only the high-quality pseudo labels and leveraging all the pseudo labels. Addressing that, we propose a novel learning approach, called Conservative-Progressive Collaborative Learning (CPCL), among which two predictive networks are trained in parallel, and the pseudo supervision is implemented based on both the agreement and disagreement of the two predictions. One network seeks common ground via intersection supervision and is supervised by the high-quality labels to ensure a more reliable supervision, while the other network reserves differences via union supervision and is supervised by all the pseudo labels to keep exploring with curiosity. Thus, the collaboration of conservative evolution and progressive exploration can be achieved. To reduce the influences of the suspicious pseudo labels, the loss is dynamic re-weighted according to the prediction confidence. Extensive experiments demonstrate that CPCL achieves state-of-the-art performance for semi-supervised semantic segmentation.

On Iterative Proportional Updating: Limitations and Improvements for General Population Synthesis.

Population synthesis is the foundation of the agent-based social simulation. Current approaches mostly consider basic population and households, rather than other social organizations. This article starts with a theoretical analysis of the iterative proportional updating (IPU) algorithm, a representative method in this field, and then gives an extension to consider more social organization types. The IPU method, for the first time, proves to be unable to converge to an optimal population distribution that simultaneously satisfies the constraints from individual and household levels. It is further improved to a bilevel optimization, which can solve such a problem and include more than one type of social organization. Numerical simulations, as well as population synthesis using actual Chinese nationwide census data, support our theoretical conclusions and indicate that our proposed bilevel optimization can both synthesize more social organization types and get more accurate results.

Bridging the Micro and Macro: Calibration of Agent-Based Model Using Mean-Field Dynamics.

Calibration of agent-based models (ABM) is an essential stage when they are applied to reproduce the actual behaviors of distributed systems. Unlike traditional methods that suffer from the repeated trial and error and slow convergence of iteration, this article proposes a new ABM calibration approach by establishing a link between agent microbehavioral parameters and systemic macro-observations. With the assumption that the agent behavior can be formulated as a high-order Markovian process, the new approach starts with a search for an optimal transfer probability through a macrostate transfer equation. Then, each agent's microparameter values are computed using mean-field approximation, where his complex dependencies with others are approximated by an expected aggregate state. To compress the agent state space, principal component analysis is also introduced to avoid high dimensions of the macrostate transfer equation. The proposed method is validated in two scenarios: 1) population evolution and 2) urban travel demand analysis. Experimental results demonstrate that compared with the machine-learning surrogate and evolutionary optimization, our method can achieve higher accuracies with much lower computational complexities.

Rod-coil block copolymer aggregates via polymerization-induced self-assembly.

Polymerization-induced self-assembly (PISA), incorporating the polymerization with in situ self-assembly, can achieve nano-objects efficiently. However, the cooperative polymerization and self-assembly lead to unclear polymerization kinetics and aggregation behavior, especially for the systems forming rigid chains. Here, we used dissipative particle dynamics simulations with a probability-based reaction model to explore the PISA behavior of rod-coil block copolymer systems. The impact of the length of macromolecular initiators, the targeted length of rigid chains, and the reaction probability on the PISA behavior, including polymerization kinetics and self-assembly, were examined. The difference between PISA and traditional self-assembly was revealed. A comparison with experimental observations shows that the simulation can capture the essential feature of the PISA. The present work provides a comprehensive understanding of rod-coil PISA systems and may provide meaningful information for future experimental research.

Biomimetic Asymmetric Polymer Brush Coatings Bearing Fencelike Conformation Exhibit Superior Protection and Antifouling Performance.

Antifouling surfaces with optimized conformation and compositional heterogeneities are presented with the goal of improving the efficacy of surface protection. The approach exploits the adhesive group (thiol or catechol chain end) to anchor asymmetric polymer brushes (APBs) bearing amphiphilic side chains with synergistic nonfouling and fouling-release abilities onto the surface. The conformation of the APB surface is close to the fencelike structure, which mimics lubricating protein lubricin, endowing the surface with capacity of enhanced protection and antiadhesivity, even facing the high compression of fouling. By utilizing a poly(Br-acrylate-alkyne) macroagent comprising alkynyl and 2-bromopropionate groups, we prepared a series of APB surfaces based on polyacrylate-g-poly(ethylene oxide)/poly(pentafluorophenyl methacrylate) (PA-g-PEO/PPFMA) APBs to explore the influence of the content of the fluorinated segment and bioinspired topological polymer chemistry on their antifouling performance. The APB surfaces can not only provide compositional heterogeneities of PEO and fluorinated segments in each side chain but also give a high surface coverage because of the characteristic of high grafting density of macromolecular brushes. It was found for the first time, as far as we are aware, the fencelike APB surface shows excellent antifouling performance with less protein adsorption (up to 91% off) and cell adhesion (up to 84% off) in comparison with the controlled substrate under relatively long incubation time.

Self-assembly and stimuli-responsive behaviours of side-chain liquid crystalline copolymers: a dissipative particle dynamics simulation approach.

We conducted a dissipative particle dynamics simulation to explore the self-assembly behaviours of side-chain liquid crystalline copolymers in solvents poor for comb blocks. Our results show that the copolymers are able to self-assemble into various aggregates such as micelles and vesicles, and especially into deformed structures such as rectangle cylinders. The morphology of the aggregates depends on the concentration of copolymers and a series of structural parameters such as the length of the coil blocks. Three kinds of morphology diagrams are mapped out to gain insight into the effect of these parameters on the self-assembly behaviours. Moreover, the aggregate structures such as the spatial arrangement of the rigid side chain are analysed, and it was found that the interfaces of the aggregates are greatly influenced by the packing of the side chains. We also investigated the influence of the rod-to-coil conformation transition of side chains on the self-assembly, motivated by stimuli-responsive polymer systems. A transformation from a disk-like structure to a vesicle was observed as the rod-to-coil conformation transition occurs. A comparison between the available experimental observations and the simulation results is made, and agreement is achieved.

Sensing Urban Transportation Events from Multi-Channel Social Signals with the Word2vec Fusion Model.

Social sensors perceive the real world through social media and online web services, which have the advantages of low cost and large coverage over traditional physical sensors. In intelligent transportation researches, sensing and analyzing such social signals provide a new path to monitor, control and optimize transportation systems. However, current research is largely focused on using single channel online social signals to extract and sense traffic information. Clearly, sensing and exploiting multi-channel social signals could effectively provide deeper understanding of traffic incidents. In this paper, we utilize cross-platform online data, i.e., Sina Weibo and News, as multi-channel social signals, then we propose a word2vec-based event fusion (WBEF) model for sensing, detecting, representing, linking and fusing urban traffic incidents. Thus, each traffic incident can be comprehensively described from multiple aspects, and finally the whole picture of unban traffic events can be obtained and visualized. The proposed WBEF architecture was trained by about 1.15 million multi-channel online data from Qingdao (a coastal city in China), and the experiments show our method surpasses the baseline model, achieving an 88.1% F1 score in urban traffic incident detection. The model also demonstrates its effectiveness in the open scenario test.

Framework for fusing traffic information from social and physical transportation data.

Tremendous volumes of messages on social media platforms provide supplementary traffic information and encapsulate crowd wisdom for solving transportation problems. However, social media messages manifested in human languages are usually characterized with redundant, fuzzy and subjective features. Here, we develop a data fusion framework to identify social media messages reporting non-recurring traffic events by connecting the traffic events with traffic states inferred from taxi global positioning system (GPS) data. Temporal-spatial information of traffic anomalies caused by the traffic events are then retrieved from anomalous traffic states. The proposed framework successfully identified accidental traffic events with various scales and exhibited strong performance in event descriptions. Even though social media messages are generally posted after the occurrence of anomalous traffic states, resourceful event descriptions in the messages are helpful in explaining traffic anomalies and for deploying suitable countermeasures.

Parallel Intelligent Systems for Integrated High-Speed Railway Operation Control and Dynamic Scheduling.

The information exchange gap between current operation control and dynamic scheduling in high-speed railway systems (HRSs) still exists, and this gap has hindered the further integrative improvement of HRSs. This paper aims to explore a feasible solution to bridging the information exchange gap for further improving the efficiency of HRSs, with the parallel intelligent systems for integrated HRS operation control and dynamic scheduling first analyzed and constructed using the ACP approach, that is, "artificial systems" (A), "computational experiments," (C) and "parallel execution" (P). Then, on the basis of the constructed parallel intelligent systems, experiments on several typical scenarios in HRSs are conducted to achieve a set of control and management strategies for actual HRSs. Experimental results show that a number of powerful tools provided by the proposed parallel intelligent systems can be utilized not only to study the current HRSs, but also to further undertake research on integrated operation control and dynamic scheduling for HRSs.

An insight into polymerization-induced self-assembly by dissipative particle dynamics simulation.

Polymerization-induced self-assembly is a one-pot route to produce concentrated dispersions of block copolymer nano-objects. Herein, dissipative particle dynamics simulations with a reaction model were employed to investigate the behaviors of polymerization-induced self-assembly. The polymerization kinetics in the polymerization-induced self-assembly were analyzed by comparing with solution polymerization. It was found that the polymerization rate enhances in the initial stage and decreases in the later stage. In addition, the effects of polymerization rate, length of macromolecular initiators, and concentration on the aggregate morphologies and formation pathway were studied. The polymerization rate and the length of the macromolecular initiators are found to have a marked influence on the pathway of the aggregate formations and the final structures. Morphology diagrams were mapped correspondingly. A comparison between simulation results and experimental findings is also made and an agreement is shown. This work can enrich our knowledge about polymerization-induced self-assembly.