Event cameras, as a bio-inspired type of dynamic visual sensors (DVS), have gained increasing attentions in the computer vision field. Existing methods for processing event data using transformers have failed to address the characteristics of small dataset size and high information density in event datasets. To overcome this issue, we propose an improved transformer-based event camera data processing framework called feedforward neural network and multilayer perceptron classifier enhanced event transformer (FM-EVT). The core innovation of this framework lies in the following aspects: Firstly, the attention separation head layer replacement technique is adopted, where the self-attention heads in the transformer are replaced with shallow neural networks to more finely analyze the information differences that each head focuses on, thereby more effectively capturing the spatial-temporal dependency relationships among event data. Secondly, focal loss is introduced to replace the traditional negative log-likelihood loss function, enabling the model to focus more on difficult-to-classify samples during training, thus enhancing the prediction performance for minority classes. In addition, by increasing the number of hidden layers and applying regularization techniques, a multilayer perceptron classifier is constructed, which enhances feature extraction and generalization capabilities, further improving classification accuracy. Experimental results on the DVS128 gesture dataset and the SL-Animals-DVS dataset demonstrate that FM-EVT excels in terms of accuracy, parameter count, and computational cost. Specifically, on the DVS128 gesture dataset, FM-EVT achieves an accuracy of 98.23%, representing a 2.03% improvement over the baseline model Event Transformer; on the SL-Animals-DVS dataset, its accuracy reaches 90%, a 1.88% increase compared to the Event Transformer model. These results fully validate the effectiveness and superiority of FM-EVT in the field of event camera data processing.

Cold-start problem is still a very challenging problem in recommender systems. Fortunately, the interactions of the cold-start users in the auxiliary source domain can help cold-start recommendations in the target domain. How to transfer user's preferences from the source domain to the target domain, is the key issue in Cross-domain Recommendation (CDR) which is a promising solution to deal with the cold-start problem. Most existing methods model a common preference bridge to transfer preferences for all users. Intuitively, since preferences vary from user to user, the preference bridges of different users should be different. Along this line, we propose a novel framework named Personalized Transfer of User Preferences for Cross-domain Recommendation (PTUPCDR). Specifically, a meta network fed with users' characteristic embeddings is learned to generate personalized bridge functions to achieve personalized transfer of preferences for each user. To learn the meta network stably, we employ a task-oriented optimization procedure. With the meta-generated personalized bridge function, the user's preference embedding in the source domain can be transformed into the target domain, and the transformed user preference embedding can be utilized as the initial embedding for the cold-start user in the target domain. Using large real-world datasets, we conduct extensive experiments to evaluate the effectiveness of PTUPCDR on both cold-start and warm-start stages. The code has been available at https://github.com/easezyc/WSDM2022-PTUPCDR.

Real-world super platforms such as Google and WeChat usually have different recommendation scenarios to provide heterogeneous items for users' diverse demands. Multi-domain recommendation (MDR) is proposed to improve all recommendation domains simultaneously, where the key point is to capture informative domain-specific features from all domains. To address this problem, we propose a novel Adversarial feature translation (AFT) model for MDR, which learns the feature translations between different domains under a generative adversarial network framework. Precisely, in the multi-domain generator, we propose a domain-specific masked encoder to highlight inter-domain feature interactions, and then aggregate these features via a transformer and a domain-specific attention. In the multi-domain discriminator, we explicitly model the relationships between item, domain and users' general/domain-specific representations with a two-step feature translation inspired by the knowledge representation learning. In experiments, we evaluate AFT on a public and an industrial MDR datasets and achieve significant improvements. We also conduct an online evaluation on a real-world MDR system. We further give detailed ablation tests and model analyses to verify the effectiveness of different components. Currently, we have deployed AFT on WeChat Top Stories. The source code is in https://github.com/xiaobocser/AFT.

In recommender systems and advertising platforms, marketers always want to deliver products, contents, or advertisements to potential audiences over media channels such as display, video, or social. Given a set of audiences or customers (seed users), the audience expansion technique (look-alike modeling) is a promising solution to identify more potential audiences, who are similar to the seed users and likely to finish the business goal of the target campaign. However, look-alike modeling faces two challenges: (1) In practice, a company could run hundreds of marketing campaigns to promote various contents within completely different categories every day, e.g., sports, politics, society. Thus, it is difficult to utilize a common method to expand audiences for all campaigns. (2) The seed set of a certain campaign could only cover limited users. Therefore, a customized approach based on such a seed set is likely to be overfitting.

More and more CDN (Content Delivery Network) or video content providers are deploying their cache nodes at the edge of networks in order to improve user perceived QoE (Quality of Experience). However, state-of-the-art ABR (Adaptive Bitrate) algorithms do not take into account the presence of a cache on video delivery path. In this paper, we firstly investigate how these ABR algorithms behave in the context of edge caching system and how caching factors such as hit ratio and access bandwidth impact on the performance of ABR algorithms. Extensive analysis results show introducing an edge cache on video delivery path cannot necessarily improve the performance of ABR algorithms in terms of average QoE. Furthermore, we propose a cache-aware ABR approach taking into account caching information including a cache hit indicator for the past chunks and the availability of the next video chunk on the cache. Experimental results validate the significant benefits of the approach which helps to make a more accurate throughput estimation and is more likely to select the video chunks stored in the cache to increase cache utilization.

Recently, deep learning models play more and more important roles in contents recommender systems. However, although the performance of recommendations is greatly improved, the "Matthew effect" becomes increasingly evident. While the head contents get more and more popular, many competitive long-tail contents are difficult to achieve timely exposure because of lacking behavior features. This issue has badly impacted the quality and diversity of recommendations. To solve this problem, look-alike algorithm is a good choice to extend audience for high quality long-tail contents. But the traditional look-alike models which widely used in online advertising are not suitable for recommender systems because of the strict requirement of both real-time and effectiveness. This paper introduces a real-time attention based look-alike model (RALM) for recommender systems, which tackles the challenge of conflict between real-time and effectiveness. RALM realizes real-time look-alike audience extension benefiting from seeds-to-user similarity prediction and improves the effectiveness through optimizing user representation learning and look-alike learning modeling. For user representation learning, we propose a novel neural network structure named attention merge layer to replace the concatenation layer, which significantly improves the expressive ability of multi-fields feature learning. On the other hand, considering the various members of seeds, we design global attention unit and local attention unit to learn robust and adaptive seeds representation with respect to a certain target user. At last, we introduce seeds clustering mechanism which not only reduces the time complexity of attention units prediction but also minimizes the loss of seeds information at the same time. According to our experiments, RALM shows superior effectiveness and performance than popular look-alike models. RALM has been successfully deployed in "Top Stories" Recommender System of WeChat, leading to great improvement on diversity and quality of recommendations. As far as we know, this is the first real-time look-alike model applied in recommender systems.

For full checkpoint on a large-scale HPC system, huge memory contexts must potentially be transferred through the network and saved in a reliable storage. As such, the time taken to checkpoint becomes a critical issue which directly impacts the total execution time. Therefore, incremental checkpoint as a less intrusive method to reduce the waste time has been gaining significant attentions in the HPC community. In this paper, we built a model that aims to reduce full checkpoint overhead by performing a set of incremental checkpoints between two consecutive full checkpoints. Moreover, a method to find the number of those incremental checkpoints is given. Furthermore, most of the comparison results between the incremental checkpoint model and the full checkpoint model (Liu et al., 2007) on the same failure data set show that the total waste time in the incremental checkpoint model is significantly smaller than the waste time in the full checkpoint model.

The increase in the physical size of High Performance Computing (HPC) platform makes system reliability more challenging. In order to minimize the performance loss (rollback and checkpoint overheads) due to unexpected failures or unnecessary overhead of fault tolerant mechanisms, we present a reliability-aware method for an optimal checkpoint/restart strategy. Our scheme aims at addressing fault tolerance challenge, especially in a large-scale HPC system, by providing optimal checkpoint placement techniques that are derived from the actual system reliability. Unlike existing checkpoint models, which can only handle Poisson failure and a constant checkpoint interval, our model can deal with a varying checkpoint interval and with different failure distributions. In addition, the approach considers optimality for both checkpoint overhead and rollback time. Our validation results suggest a significant improvement over existing techniques.

The increase in the physical size of High Performance Computing (HPC) platform makes system reliability more challenging. In order to minimize the performance loss due to unexpected failures or unnecessary overhead of fault tolerant mechanisms, we present a reliability-aware method for an optimal checkpoint/restart strategy towards minimizing rollback and checkpoint overheads. Our scheme aims to address fault tolerance challenge especially in a large-scale HPC system by providing optimal checkpoint placement techniques that are derived from the actual system reliability. Unlike existing checkpoint models, which can only handle Poisson failure and a constant checkpoint interval, our model can perform a varying checkpoint interval and deal with different failure distributions. In addition, the approach considers optimality for both checkpoint overhead and rollback time. Our validation results suggest a significant improvement over existing techniques.

In recent years, large scale clusters have been commonly deployed to solve important grand-challenge scientific problems. In order to reduce computational time, the system size has been increasingly expanded. Unfortunately, the reliability of such cluster systems goes in the opposite direction, as the extension of a system scale. Since failures of a single node could result in a system outage, it is essential to effectively deal with faulty situations in the grand challenge problem-solving environment. Checkpointing is one of common fault tolerance techniques. However, there are many challenges in checkpointing such as overhead, latency and consistency, as well as recovery. In this paper, a reliability-aware checkpoint/restart method was introduced. It is a novel technique to consider checkpointing placement based on system reliability. We constructed a cost model and derived an optimal checkpoint placement function based on failure rates: A trade-off between performance and reliability (i.e. performability) was a key consideration. We also implemented a proof-of-concept and demonstrated improvements resulting from our techniques for fault-tolerant MPI applications on an HA-OSCAR cluster.