Sentiment and emotion, which correspond to long-term and short-lived human feelings, are closely linked to each other, leading to the fact that sentiment analysis and emotion recognition are also two interdependent tasks in natural language processing (NLP). One task often leverages the shared knowledge from another task and performs better when solved in a joint learning paradigm. Conversational context dependency, multi-modal interaction, and multi-task correlation are three key factors that contribute to this joint paradigm. However, none of the recent approaches have considered them in a unified framework. To fill this gap, we propose a multi-modal, multi-task interactive graph attention network, termed M3GAT, to simultaneously solve the three problems. At the heart of the model is a proposed interactive conversation graph layer containing three core sub-modules, which are: (1) local-global context connection for modeling both local and global conversational context, (2) cross-modal connection for learning multi-modal complementary and (3) cross-task connection for capturing the correlation across two tasks. Comprehensive experiments on three benchmarking datasets, MELD, MEISD, and MSED, show the effectiveness of M3GAT over state-of-the-art baselines with the margin of 1.88%, 5.37%, and 0.19% for sentiment analysis, and 1.99%, 3.65%, and 0.13% for emotion recognition, respectively. In addition, we also show the superiority of multi-task learning over the single-task framework.

Sarcasm is a form of figurative language device to express human inner feelings, where the author writes the positive sentence on surface form, while he/she actually expresses negative sentiment, vice versa. Sentiment thus comes into sight, and is closely related with sarcasm, leading to the recent popularity of multi-modal sarcasm and sentiment joint detection in conversation (dialogue). The key challenges involve multi-modal fusion and multi-task interaction. Most of the existing studies have focused on building multi-modal fused representation, while the commonness and uniqueness across related tasks has not received attention. To fill this gap, we propose a multi-modal multi-task interaction learning framework, termed MIL, for joint detection of sarcasm and sentiment. Specifically, a cross-modal target attention mechanism is proposed to automatically learn the alignment between texts and images/speeches. In addition, a multi-modal interaction learning paradigm consisting of a dual-gating network, three separate fully-connected layers that simultaneously capture the commonness and uniqueness. Comprehensive experiments on two benchmarking datasets (i.e., Memotion and MUStARD) show the effectiveness of the proposed model over state-of-the-art baselines with a significant improvement of 1.9%, 2.4% in terms of F1.

Temporal link prediction, aiming to predict future edges between paired nodes in a dynamic graph, is of vital importance in diverse applications. However, existing methods are mainly built upon uniform Euclidean space, which has been found to be conflict with the power-law distributions of real-world graphs and unable to represent the hierarchical connections between nodes effectively. With respect to the special data characteristic, hyperbolic geometry offers an ideal alternative due to its exponential expansion property. In this paper, we propose HGWaveNet, a novel hyperbolic graph neural network that fully exploits the fitness between hyperbolic spaces and data distributions for temporal link prediction. Specifically, we design two key modules to learn the spatial topological structures and temporal evolutionary information separately. On the one hand, a hyperbolic diffusion graph convolution (HDGC) module effectively aggregates information from a wider range of neighbors. On the other hand, the internal order of causal correlation between historical states is captured by hyperbolic dilated causal convolution (HDCC) modules. The whole model is built upon the hyperbolic spaces to preserve the hierarchical structural information in the entire data flow. To prove the superiority of HGWaveNet, extensive experiments are conducted on six real-world graph datasets and the results show a relative improvement by up to 6.67% on AUC for temporal link prediction over SOTA methods.

The intelligent patrol car with environmental sensing and autonomous navigation is a special robot, which is mostly used for equipment defect detection in industrial areas such as the power distribution room or data center room. A path planning algorithm for the navigation system of intelligent patrol car is proposed to ensure efficient and secure navigation in the complex indoor environment, and its effect is verified by simulation and experiment. First, a patrol car platform integrated with several intelligent devices is built to achieve global localization, mapping and path planning. Then a new co-optimization on multi-objective Cauchy mutation cat swarm optimization (MOCMCSO) and artificial potential field method (APFM) is proposed to solve the multi-objective optimization problem on shortest global path length and minimum total turning-angle variation. The optimal path is written into the navigation module to drive the patrol car to move and navigate. The simulations are carried out to confirm that the method can achieve a balance between the shortest path and good path smoothness, which has less optimization time and lower fitness value compared with multi-objective cat swarm optimization (MOCSO) and multi-objective particle swarm optimization (MOPSO), and is more suitable for global path planning in indoor environment. Finally, the experiments have been carried out in the data center equipment room to verify the effectiveness and superiority over the path planning algorithm on MOCMCSO.

Most of image quality assessment methods only concern about gray image, and don't make use of image color information sufficiently at present. A method for reduced-reference color image quality assessment is proposed here, which efficiently uses the color information, performs the spatial domain color difference analysis for components H and S, conducts frequency domain distortion analysis for component I transformed by Contourlet. The final image quality assessment is defined by the weighted combination of three components. The experiments show that the proposed objective method has a good coincidence with the subjective perception, and can reflect the image quality effectively.

The authors study a number of non-uniform specified topologies and show rigorously that for certain topologies with cut edges, the critical exponent gamma _t is in agreement with a conjecture given by Gaunt et al. (1984) and that the exponent nu _t= nu , the exponent for self-avoiding walks. The authors also find that the scaling relations gamma _t- gamma _t¹ and gamma _t- gamma _t¹¹ are the same as for self-avoiding walks, previously conjectured only for uniform networks. By assigning an interaction energy to a nearest neighbour contact, they prove that the collapse transition for these topologies is the same as that for self-avoiding walks.