Clothing is one of the symbols of human civilisation. Clothing design is an art form that combines practicality and artistry. The Dunhuang clothes culture has a long history which represents ancient Chinese aesthetics. Artificial intelligence (AI) technology has been recently applied to multiple areas, which is also drawing increasing attention in fashion. However, little research has been done on the usage of AI for the creation of clothing, especially in traditional culture. It is challenging that the exploration of computer science and Dunhuang clothing design, which is a cross-history interaction between AI and Chinese classical culture. In this paper, we propose ClothGAN, which is an innovative framework for “designing” new patterns and styles of clothes based on generative adversarial network (GAN) and style transfer algorithm. Besides, we built the Dunhuang clothes dataset and conducted experiments to generate new patterns and styles of clothes with Dunhuang elements. We evaluated these clothing works generated from different models by computing inception score (IS), human prefer score (HPS) and generated score (IS and HPS). The results show that our framework outperformed others in these designing works.

Traffic flow prediction, one of the essential problems in traffic control and guidance systems, is still drawing increasing attention in recent years with the new methods tipped by the success of AI. In this paper, we propose a novel model, namely self-attention generative adversarial networks for time-series prediction (SATP-GAN). The SATP-GAN method is based on self-attention and generative adversarial networks (GAN) mechanisms, which are composed of the GAN module and reinforcement learning (RL) module. In the GAN module, we apply the self-attention layer to capture the pattern of time-series data instead of RNNs (recurrent neural networks). In the RL module, we apply the RL algorithm to tune the parameters of our SATP-GAN model. We evaluate the framework on the real-world traffic dataset and obtain a consistent improvement of 6.5% over baseline methods. The SATP-GAN framework proves the GAN mechanism is also available for time-series prediction after fine-tuning the parameters.

Travel time prediction is critical for advanced traveler information systems (ATISs), which provides valuable information for enhancing the efficiency and effectiveness of the urban transportation systems. However, in the area of bus trips, existing studies have focused on directly using the structured data to predict travel time for a single bus trip. For state-of-the-art public transportation information systems, a bus journey generally has multiple bus trips. Additionally, due to the lack of study on data fusion, it is even inadequate for the development of underlying intelligent transportation systems. In this paper, we propose a novel framework for a hybrid data-driven travel time prediction model for bus journeys based on open data. We explore a convolutional long short-term memory (ConvLSTM) model with a self-attention mechanism that accurately predicts the running time of each segment of the trips and the waiting time at each station. The model is more robust to capture long-range dependence in time series data as well.

With smart city infrastructures growing, the Internet of Things (IoT) has been widely used in the intelligent transportation systems (ITS). The traditional adaptive traffic signal control method based on reinforcement learning (RL) has expanded from one intersection to multiple intersections. In this paper, we propose a multi-agent auto communication (MAAC) algorithm, which is an innovative adaptive global traffic light control method based on multi-agent reinforcement learning (MARL) and an auto communication protocol in edge computing architecture. The MAAC algorithm combines multi-agent auto communication protocol with MARL, allowing an agent to communicate the learned strategies with others for achieving global optimization in traffic signal control. In addition, we present a practicable edge computing architecture for industrial deployment on IoT, considering the limitations of the capabilities of network transmission bandwidth. We demonstrate that our algorithm outperforms other methods over 17% in experiments in a real traffic simulation environment.

Traffic signal control (TSC) is still one of the most significant and challenging research problems in the transportation field. Reinforcement learning (RL) has achieved great success in TSC but suffers from critically high learning costs in practical applications due to the excessive trial-and-error learning process. Offline RL is a promising method to reduce learning costs whereas the data distribution shift issue is still up in the air. To this end, in this paper, we formulate TSC as a sequence modeling problem with a sequence of Markov decision process described by states, actions, and rewards from the traffic environment. A novel framework, namely TransformerLight, is introduced, which does not aim to fit into value functions by averaging all possible returns, but produces the best possible actions using a gated Transformer. Additionally, the learning process of TransformerLight is much more stable by replacing the residual connections with gated transformer blocks due to a dynamic system perspective. Through numerical experiments on offline datasets, we demonstrate that the TransformerLight model: (1) can build a high-performance adaptive TSC model without dynamic programming; (2) achieves a new state-of-the-art compared to most published offline RL methods so far; and (3) shows a more stable learning process than offline RL and recent Transformer-based methods. The relevant dataset and code are available at Github.