In recommender systems, the patterns of user behaviors (e.g., purchase, click) may vary greatly in different contexts (e.g., time and location). This is because user behavior is jointly determined by two types of factors: intrinsic factors , which reflect consistent user preference, and extrinsic factors , which reflect external incentives that may vary in different contexts. Differentiating between intrinsic and extrinsic factors helps learn user behaviors better. However, existing studies have only considered differentiating them from a single, pre-defined context (e.g., time or location), ignoring the fact that a user’s extrinsic factors may be influenced by the interplay of various contexts at the same time. In this article, we propose the intrinsic-extrinsic disentangled recommendation (IEDR) model, a generic framework that differentiates intrinsic from extrinsic factors considering various contexts simultaneously, enabling more accurate differentiation of factors and hence the improvement of recommendation accuracy. IEDR contains a context-invariant contrastive learning component to capture intrinsic factors, and a disentanglement component to extract extrinsic factors under the interplay of various contexts. The two components work together to achieve effective factor learning. Extensive experiments on real-world datasets demonstrate IEDR’s effectiveness in learning disentangled factors and significantly improving recommendation accuracy by up to 4% in NDCG.

Aggregating distributed energy resources in power systems significantly increases uncertainties, in particular caused by the fluctuation of renewable energy generation. This issue has driven the necessity of widely exploiting advanced predictive control techniques under uncertainty to ensure long-term economics and decarbonization. In this paper, we propose a real-time uncertainty-aware energy dispatch framework, which is composed of two key elements: (i) A hybrid forecast-and-optimize sequential task, integrating deep learning-based forecasting and stochastic optimization, where these two stages are connected by the uncertainty estimation at multiple temporal resolutions; (ii) An efficient online data augmentation scheme, jointly involving model pre-training and online fine-tuning stages. In this way, the proposed framework is capable to rapidly adapt to the real-time data distribution, as well as to target on uncertainties caused by data drift, model discrepancy and environment perturbations in the control process, and finally to realize an optimal and robust dispatch solution. The proposed framework won the championship in CityLearn Challenge 2022, which provided an influential opportunity to investigate the potential of AI application in the energy domain. In addition, comprehensive experiments are conducted to interpret its effectiveness in the real-life scenario of smart building energy management.

With rapid evolution of e-commerce, it is essential but challenging to quickly provide a recommending service for users. The recommender system can be divided into two stages: retrieval and ranking. However, most recent academic research has focused on the second stage for datasets with limited size, while the role of retrieval is heavily underestimated. Generally, graph-based or sequential models are used to generate item embedding for the retrieval task. However, the graph-based methods suffer from over-smoothing, while sequential models are largely influenced by data sparseness. To alleviate these issues, we propose NATM—a novel embedding-based method in large-scale learning incorporating both graph-based and sequential information. NATM consists of two key components: i) neighbor augmented graph construction with user behaviors to enhance item embedding and mitigate data sparseness, followed by ii) transformer-based representation network, targeting on minimizing NCE loss. The competitive performance of the proposed method is demonstrated through comprehensive experiments, including a benchmark study on MovieLens dataset and a real-world e-commerce scenario in Alibaba Group.

Sequential prediction is one of the key components in recommendation. In online e-commerce recommendation system, user behavior consists of the sequential visiting logs and item behavior contains the interacted user list in order. Most of the existing state-of-the-art sequential prediction methods only consider the user behavior while ignoring the item behavior. In addition, we find that user behavior varies greatly at different time, and most existing models fail to characterize the rich temporal information. To address the above problems, we propose a transformer-based spatial-temporal recommendation framework (STEM). In the STEM framework, we first utilize attention mechanisms to model user behavior and item behavior, and then exploit spatial and temporal information through a transformer-based model. The STEM framework, as a plug-in, is able to be incorporated into many neural network-based sequential recommendation methods to improve performance. We conduct extensive experiments on three real-world Amazon datasets. The results demonstrate the effectiveness of our proposed framework.

With the development of deep learning techniques, contrastive representation learning has been increasingly employed in large-scale recommender systems. For instance, deep user-item matching models can be trained by contrasting positive and negative examples and learning discriminative user and item representations. Despite their success, the distinguishable properties of the recommender system are often ignored in existing modelling. Standard methods approximate maximum likelihood estimation on user behavior data in a manner similar to language models. Specifically, the way of model optimization corresponds to approximating the user-item pointwise mutual information, which can be regarded as eliminating the influence of global item popularity on user behavior to capture intrinsic user preference. In addition, unlike the situation in language models where word frequency is relatively stable, item popularity is constantly evolving. To address these issues, we propose a novel dynamic popularity-aware (DPA) contrastive learning method for recommendation, which consists of two key components: i) a dynamic negative sampling strategy is involved to enhance the user representation, ii) a dynamic prediction recovery is adopted by the real-time item popularity. The proposed strategy can be naturally overlaid on any contrastive learning-based matching model to more accurately capture user interest and system dynamics. Finally, the effectiveness of the proposed strategy is demonstrated through comprehensive experiments on an e-commerce scenario of Alibaba Group.