A Contrastive Hypergraph Transformer for session-based recommendation (CHT) model was proposed to address the problems of noise interference and sample sparsity in the session-based recommendation itself. Firstly, the session sequence was modeled as a hypergraph. Secondly, the global context information and local context information of items were constructed by the hypergraph transformer. Finally, the Item-Level (I-L) encoder and Session-Level (S-L) encoder were used on global relationship learning to capture different levels of item embeddings, the information fusion module was used to fuse item embedding and reverse position embedding, and the global session representation was obtained by the soft attention module while the local session representation was generated with the help of the weight line graph convolutional network on local relationship learning. In addition, a contrastive learning paradigm was introduced to maximize the mutual information between the global and local session representations to improve the recommendation performance. Experimental results on several real datasets show that the recommendation performance of CHT model is better than that of the current mainstream models. Compared with the suboptimal model S2-DHCN (Self-Supervised Hypergraph Convolutional Networks), the proposed model has the P@20 of 35.61% and MRR@20 of 17.11% on Tmall dataset, which are improved by 13.34% and 13.69% respectively; the P@20 reached 54.07% and MRR@20 reached 18.59% on Diginetica dataset, which are improved by 0.76% and 0.43% respectively; verifying the effectiveness of the proposed model.
The purpose of disentangled representation learning is to model the key factors that affect the form of data, so that the change of a key factor only causes the change of data on a certain feature, while the other features are not affected. It is conducive to face the challenge of machine learning in model interpretability, object generation and operation, zero-shot learning and other issues. Therefore, disentangled representation learning always be a research hotspot in the field of machine learning. Starting from the history and motives of disentangled representation learning, the research status and applications of disentangled representation learning were summarized, the invariance, reusability and other characteristics of disentangled representation learning were analyzed, and the research on the factors of variation via generative entangling, the research on the factors of variation with manifold interaction, and the research on the factors of variation using adversarial training were introduced, as well as the latest research trends such as a Variational Auto-Encoder (VAE) named β-VAE were introduced. At the same time, the typical applications of disentangled representation learning were shown, and the future research directions were prospected.
According to partially known probability distribution of demand information on the production-distribution network of perishable products, WCVaR (Worst-Case Conditional Value-at-Risk) was introduced to measure the risk. On the basis of considering the effect of factors, such as production, logistics distribution, transportation path etc, on production cost, transportation cost, storage cost and loss of stockout, an optimization model with minimum WCVaR at certain service level was proposed. And then the best optimization strategy was realized by minimizing tail risk loss of production-distribution network. The numerical simulation results show that the WCVaR method can handle the uncertainty with more volatility and has more excellent stability, compared with the robust optimization method. When the demand obeys mixed distribution, the optimization problem of production-distribution network with uncertainty can be well solved with WCVaR optimization model.
Now the integer Discrete Cosine Transform (DCT) algorithm of H.264 can not apply to Distributed Video Coding (DVC) framework directly because of its high complexity. In view of this, the authors presented a integer DCT algorithm and transform radix generating method based on fixed long step quantization which length was 2x (x was a plus integer). The transform radix in H.264 could be stretched. The authors took full advantage of this feature to find transform radix which best suits for working principle of hardware, and it moved the contracted-quantized stage from coder to decoder to reduced complexity of coder under the premise of "small" transform radix. In the process of "moving", this algorithm guaranteed image quality by saturated amplification for DCT coefficient, guaranteed reliability by overflow upper limit, and improved compression performance by reducing radix error. The experimental results show that, compared with corresponding module in H.264, the quantization method of this algorithm is convenient for bit-plane extraction. And it reduces calculating work of contracted-quantized stage of coder to 16 times of integer constant addition under the premise of quasi-lossless compression, raises the ratio of image quality and compression by 0.239. This algorithm conforms to DVC framework.