The problem of adaptive traffic signal control in multi-intersection system has attracted the attention of researchers. Among the existing methods, the reinforcement learning has shown to be effective. However, the complex intersection features, heterogeneous intersection structures and dynamic coordination for multiple intersections pose challenges for reinforcement learning based algorithms. This paper proposes a cooperative deep Q-network with Q-value transfer (QT-CDQN) for adaptive multiintersection signal control. In QT-CDQN, a multi-intersection traffic network in a region is modeled as a multi-agent reinforcement learning system. Each agent searches the optimal strategy to control an intersection by a deep Q-network that takes the discrete state encoding of traffic information as the network inputs. To work cooperatively, the agent considers the influence of the latest actions of its adjacencies in the process of policy learning. Specially, the optimal Q-values of the neighbor agents at the latest time step are transferred to the loss function of the Q-network. Moreover, the strategy of target network and the mechanism of experience replay are used to improve the stability of the algorithm. The advantages of QTCDQN lie not only in the effectiveness and scalability for multi-intersection system, but also in the versatility to deal with heterogeneous intersection structures. Experimental studies under different road structures show that the QT-CDQN is competitive in terms of average queue length, average speed and average waiting time when compared with the state-of-the-art algorithms. Furthermore, the experiments of recurring congestion and occasional congestion validate the adaptability of the QT-CDQN to dynamic traffic environments.

Researches have shown difficulties in obtaining proximity while maintaining diversity for many-objective optimization problems (MaOPs). Complexities of the true Pareto Front (PF) pose challenges for the reference vector based algorithms for their insufficient adaptability to the diverse characteristics with no priori. This paper proposes a manyobjective optimization Algorithm with two Interacting processes: cascade Clustering and reference point incremental Learning (CLIA). In the population selection process based on cascade clustering, using the reference vectors provided by the process based on incremental learning, the non-dominated and the dominated individuals are clustered and sorted with different manners in a cascade style and are selected by round-robin for better proximity and diversity. In the reference vector adaptation process based on reference point incremental learning, using the feedbacks from the process based on cascade clustering, proper distribution of reference points is gradually obtained by incremental learning. Experimental studies on several benchmark problems show that CLIA is competitive compared with the stateof-the-art algorithms and has impressive efficiency and versatility using only the interactions between the two processes without incurring extra evaluations.

Multi-agent reinforcement learning (MARL) based methods for adaptive traffic signal control (ATSC) have shown promising potentials to solve the heavy traffic problems. The existing MARL methods adopt centralized or distributed strategies. The former only models the environment as an agent and suffers from the exponential growth of action and state space. The latter extends the independent reinforcement learning methods, such as DQN, to multiple interactions directly or propagates information, such as state and policy, without taking their qualities into account. In this paper, we propose a multi-agent transfer reinforcement learning method to enhance the performance of MARL for ATSC, which is termed as multi-agent transfer soft actor-critic with the multi-view encoder (MT-SAC). The MT-SAC combines centralized and distributed strategies. In MT-SAC, we propose a multi-view state encoder and a transfer learning paradigm with guidance. The encoder processes input states from multiple perspectives and uses an attention mechanism to weigh the neighborhood information. While the paradigm enables the agents to handle different conditions for improving generalization abilities by transfer learning. Experimental studies on different scale road networks show that the MT-SAC outperforms the state-of-the-art algorithms and makes the traffic signal controllers more collaborative and robust.

Video action recognition is a fundamental task for video understanding. Action recognition in complex spatio-temporal contexts generally requires fusing of different multi-granularity action information. However, existing works do not consider spatio-temporal information modeling and fusion from the perspective of action granularity. To address this problem, this paper proposes an Action Granularity Pyramid Network (AGPN) for action recognition, which can be flexibly integrated into 2D backbone networks. The core module is the Action Granularity Pyramid Module (AGPM), a hierarchical pyramid structure with residual connections, which is established to fuse multi-granularity action spatio-temporal information. From top to bottom level in the designed pyramid structure, the receptive field decreases and action granularity becomes more refined. To enrich temporal information of the inputs, a Multiple Frame Rate Module (MFM) is proposed to mix different frame rates at a fine-grained pixel-wise level. Moreover, a Spatio-temporal Anchor Module (SAM) is employed to fix spatio-temporal feature anchors to promote the effectiveness of feature extraction. We conduct extensive experiments on three large-scale action recognition datasets, Something-Something V1 & V2 and Kinetics-400. The results demonstrate that our proposed AGPN outperforms the state-of-the-art methods for the tasks of video action recognition.

Human motion prediction intends to predict how humans move given a historical sequence of 3D human motions. Recent transformer-based methods have attracted increasing attentions and demonstrated their promising performance in 3D human motion prediction. However, existing methods generally decompose the input of human motion information into spatial and temporal branches in a separate way and seldom consider their inherent coherence between the two branches, hence often failing to register the dynamic spatio-temporal information during the training process. Motivated by these issues, we propose a spatio-temporal cross-transformer network (STCT) for 3D human motion predictions. Specifically, we investigate various types of interaction methods (i.e., Concatenation Interaction, Msg token interaction, and Cross-transformer) to capture the coherence of the spatial and temporal branches. According to the obtained results, the proposed cross-transformer interaction method shows its superiority over other methods. Meanwhile, considering that most existing works treat the human body as a set of 3D human joint positions, the predicted human joints are proportionally less appropriate to the realistic human body due to unreasonable bone length and non-plausible poses as time progresses. We further resort to the bone constraints of human mesh to produce more realistic human motions. By fitting a parametric body model (i.e., SMPL-X model) to the predicted human joints, a reconstruction loss function is proposed to remedy the unreasonable bone length and pose errors. Comprehensive experiments on AMASS and Human3.6M datasets have demonstrated that our method achieves superior performance over compared methods.