In cooperative multi-agent reinforcement learning (MARL), value function factorization methods have been proposed to address the dimensionality explosion problem encountered in centralized training and decentralized execution (CTDE). The existing value function factorization methods lack a perspective from the global system when addressing credit assignment, failing to measure each agent’s contribution to the current system. A typical limitation is that these factorization approaches primarily rely on the proximity of local individual groups to assign credit. There are challenges such as the sensitivity of credit assignment weights to local information and instability in the learning process. In this study, we focus on redistributing agents’ features during the system evolution and suggest employing an enhanced attention mechanism with system entropy measure to factorize value function. Specifically, this method emphasizes each agent’s representation between their local and global contributions and then redesigns multi-head attention to optimize the credit assignment in value function factorization. To evaluate the effectiveness of our method, we conduct a series of cooperative multi-agent tasks on the StarCraft II platform and compare the results with several state-of-the-art techniques, including Qatten, QMIX, QTRAN, COMA, and VDN. The experimental results demonstrate that our method achieves faster overall convergence speed, higher stability, and robust performance across various scenarios.

Volatility forecasting is a problem in finance that attracts the attention of both academia and industry. While existing approaches typically utilize a discrete-time latent process that governs the volatility to forecast its future level, volatility is considered to evolve continuously, which makes discrete-time modeling inevitably lose some critical information about the evolution of volatility. In this article, a novel neural-network-based model, Continuous Volatility Forecasting Model, CVFM is proposed to tackle this problem. First, CVFM introduces a continuous-time latent process, whose evolution is modeled with neural differential equations (NDEs), to govern volatility, which effectively captures the continuous evolutionary behavior of volatility in a data-driven way. Second, a scale-similarity-based mechanism is designed to calibrate the evolution equation of the latent process with real-world observations in the absence of high-frequency data. CVFM is tested on six real-world stock index datasets. The main experimental results show that CVFM can significantly outperform existing models in terms of both forecasting accuracy and high-volatility recognition.

Time series forecasting has attracted significant interest across various fields in recent years. Notably, Transformers have been extensively investigated for long-term time series forecasting (LTSF) due to their remarkable ability on modeling sequential data. However, the point-wise calculation of its self-attention leads to a challenging task for accurately capturing real-world time series' local and global characteristics, especially with multiple seasonal periodic components and outliers. In this article, we leverage wavelet analysis to recognize different frequency patterns and design an effective attention mechanism for time series forecasting to address this issue. In detail, we employ the maximal overlap discrete wavelet transform (MODWT) to construct a novel wavelet attention (WA) mechanism and propose the wavelet transformer (Waveformer) prediction technique. This approach effectively extracts multiple periodic features, mitigates the influence of anomalies and improves the precision of time series prediction under seasonal-trend decomposition methods. Experimental evaluations on six real-world datasets from various application fields demonstrate that the multiple periodic decomposition strategy of Waveformer successfully captures time series seasonal patterns and improves forecasting performance in comparison with many state-of-art methods.

Graph embedding provides a feasible methodology to conduct pattern classification for graph-structured data by mapping each data into the vectorial space. Various pioneering works are essentially coding method that concentrates on a vectorial representation about the inner properties of a graph in terms of the topological constitution, node attributions, link relations, etc. However, the classification for each targeted data is a qualitative issue based on understanding the overall discrepancies within the dataset scale. From the statistical point of view, these discrepancies manifest a metric distribution over the dataset scale if the distance metric is adopted to measure the pairwise similarity or dissimilarity. Therefore, we present a novel embedding strategy named$\mathbf {MetricDistribution2vec}$to extract such distribution characteristics into the vectorial representation for each data. We demonstrate the application and effectiveness of our representation method in the supervised prediction tasks on extensive real-world structural graph datasets. The results have gained some unexpected increases compared with a surge of baselines on all the datasets, even if we take the lightweight models as classifiers. Moreover, the proposed method also conducts investigations in Few-Shot classification scenarios, and the results still show attractive discrimination in rare training samples based inference.

Molecular fingerprints are significant cheminformatics tools to map molecules into vectorial space according to their characteristics in diverse functional groups, atom sequences, and other topological structures. In this paper, we investigate a novel molecular fingerprint Anonymous-FP that possesses abundant perception about the underlying interactions shaped in small, medium, and large-scale atom chains. In detail, the possible atom chains from each molecule are sampled and extended as anonymous atom chains using an anonymous encoding manner. After that, the molecular fingerprint Anonymous-FP is embedded into vectorial space in virtue of the Natural Language Processing technique PV-DBOW. Anonymous-FP is studied on molecular property identification via molecule classification experiments on a series of molecule databases and has shown valuable advantages such as less dependence on prior knowledge, rich information content, full structural significance, and high experimental performance. During the experimental verification, the scale of the atom chain or its anonymous pattern is found significant to the overall representation ability of Anonymous-FP. Generally, the typical scale r = 8 could enhance the molecule classification performance, and specifically, Anonymous-FP gains the classification accuracy to above 93% on all NCI datasets.

Nowadays, graph structure data has played a key role in machine learning because of its simple topological structure, and therefore, the graph representation learning methods have attracted great attention. And it turns out that the low-dimensional embedding representation obtained by graph representation learning is extremely useful in various typical tasks, such as node classification and content recommendation. However, most of the existing methods do not further dig out potential structural information on the original graph structure. Here, we propose wGCN, which utilizes random walk to obtain the node-specific mesoscopic structures (high-order local structure) of the graph and utilizes these mesoscopic structures to enhance the graph and organize the characteristic information of the nodes. Our method can effectively generate node embedding for data of previously unknown categories, which has been proven in a series of experiments conducted on many types of graph networks. And compared to baselines, our method shows the best performance on most datasets and achieves competitive results on others. It is believed that combining the mesoscopic structure to further explore the structural information of the graph will greatly improve the learning efficiency of the graph neural network.

Packet classification is a key function for software-defined networking switches. For example, OpenFlow Switch examines up to 15 required fields, against thousands of rules. With the proliferation of new header fields in a ruleset, it becomes a great challenge to design a high performance packet classification solution. In this paper, we present a scalable many-field packet classification algorithm and its prototype implementation on a graphics processing unit (GPU). By exploiting the sparsity of ruleset, our algorithm uses a few effective bits (EB) to divide a large ruleset into multiple subsets with low rule replication for economic memory usage at the offline stage. These EB are chosen based on our selection metrics: wildcard ratio, independence index, and diversity index. Using these EB, our algorithm can quickly filter out the majority rules which do not need to be full-matched to improve the online system performance. Moreover, the choice of EB is adjustable to meet the implementation requirements of varying environments for good performance scalability. Our prototype on a single NVIDIA K20C GPU achieves more than 160 MPPS throughput for 100K 15-field synthetic ruleset.