Over the past 10 years, many recommendation techniques have been based on embedding users and items in latent vector spaces, where the inner product of a (user,item) pair of vectors represents the predicted affinity of the user to the item. A wealth of literature has focused on the various modeling approaches that result in embeddings, and has compared their quality metrics, learning complexity, etc. However, much less attention has been devoted to the issues surrounding productization of an embeddings-based high throughput, low latency recommender system. In particular, how the system might keep up with the changing embeddings as new models are learnt. This paper describes a reference architecture of a high-throughput, large scale recommendation service which leverages a search engine as its runtime core. We describe how the search index and the query builder adapt to changes in the embeddings, which often happen at a different cadence than index builds. We provide solutions for both id-based and feature-based embeddings, as well as for batch indexing and incremental indexing setups. The described system is at the core of a Web content discovery service that serves tens of billions recommendations per day in response to billions of user requests.
Deep learning recommendation models have grown to the terabyte scale. Traditional serving schemes--that load entire models to a single server--are unable to support this scale. One approach to support this scale is with distributed serving, or distributed inference, which divides the memory requirements of a single large model across multiple servers. This work is a first-step for the systems research community to develop novel model-serving solutions, given the huge system design space. Large-scale deep recommender systems are a novel workload and vital to study, as they consume up to 79% of …
Learned embeddings for products are an important building block for web-scale e-commerce recommendation systems. At Pinterest, we build a single set of product embeddings called ItemSage to provide relevant recommendations in all shopping use cases including user, image and search based recommendations. This approach has led to significant improvements in engagement and conversion metrics, while reducing both infrastructure and maintenance cost. While most prior work focuses on building product embeddings from features coming from a single modality, we introduce a transformer-based architectur…
Fudan University（复旦大学）；Shanghai Institute of Intelligent Electronics & Systems（上海智能电子与系统研究院）
Most of existing embedding based recommendation models use embeddings (vectors) corresponding to a single fixed point in low-dimensional space, to represent users and items. Such embeddings fail to precisely represent the users/items with uncertainty often observed in recommender systems. Addressing this problem, we propose a unified deep recommendation framework employing Gaussian embeddings, which are proven adaptive to uncertain preferences exhibited by some users, resulting in better user representations and recommendation performance. Furthermore, our framework adopts Monte-Carlo sampling…
Recommender systems are an essential component of e-commerce marketplaces, helping consumers navigate massive amounts of inventory and find what they need or love. In this paper, we present an approach for generating personalized item recommendations in an e-commerce marketplace by learning to embed items and users in the same vector space. In order to alleviate the considerable cold-start problem present in large marketplaces, item and user embeddings are computed using content features and multi-modal onsite user activity respectively. Data ablation is incorporated into the offline model tra…
Session based model is widely used in recommend system. It use the user click sequence as input of a Recurrent Neural Network (RNN), and get the output of the RNN network as the vector embedding of the session, and use the inner product of the vector embedding of session and the vector embedding of the next item as the score that is the metric of the interest to the next item. This method can be used for the "match" stage for the recommendation system whose item number is very big by using some index method like KD-Tree or Ball-Tree and etc.. But this method repudiate the variousness of the in…
Aristotle University of Thessaloniki（塞萨洛尼基亚里士多德大学）；Open University of Cyprus（塞浦路斯开放大学）
The rapid growth of users' involvement in Location-Based Social Networks (LBSNs) has led to the expeditious growth of the data on a global scale. The need of accessing and retrieving relevant information close to users' preferences is an open problem which continuously raises new challenges for recommendation systems. The exploitation of Points-of-Interest (POIs) recommendation by existing models is inadequate due to the sparsity and the cold start problems. To overcome these problems many models were proposed in the literature, but most of them ignore important factors such as: geographical p…
School of Computer Science and Engineering, Nanyang Technological University（南洋理工大学计算机科学与工程学院）；Joint NTU-UBC Research Centre of Excellence in Active Living for the Elderly (LILY)（南洋理工大学-不列颠哥伦比亚大学老年活跃生活联合卓越研究中心）；Alibaba-NTU Singapore Joint Research Institute（阿里巴巴-南洋理工大学新加坡联合研究院）；Alibaba Group（阿里巴巴集团）
Proper initialization is crucial to the optimization and the generalization of neural networks. However, most existing neural recommendation systems initialize the user and item embeddings randomly. In this work, we propose a new initialization scheme for user and item embeddings called Laplacian Eigenmaps with Popularity-based Regularization for Isolated Data (LEPORID). LEPORID endows the embeddings with information regarding multi-scale neighborhood structures on the data manifold and performs adaptive regularization to compensate for high embedding variance on the tail of the data distribut…
Do Not Wait: Learning Re-Ranking Model Without User Feedback At Serving Time in E-Commerce
Alibaba Group（阿里巴巴集团）；Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）；School of Information, Renmin University of China（中国人民大学信息学院）
Modern deep neural networks (DNNs) have greatly facilitated the development of sequential recommender systems by achieving state-of-the-art recommendation performance on various sequential recommendation tasks. Given a sequence of interacted items, existing DNN-based sequential recommenders commonly embed each item into a unique vector to support subsequent computations of the user interest. However, due to the potentially large number of items, the over-parameterised item embedding matrix of a sequential recommender has become a memory bottleneck for efficient deployment in resource-constrain…
University of Electronic Science and Technology of China（电子科技大学）；University College London（伦敦大学学院）；Tsinghua University（清华大学）
The embedding-based representation learning is commonly used in deep learning recommendation models to map the raw sparse features to dense vectors. The traditional embedding manner that assigns a uniform size to all features has two issues. First, the numerous features inevitably lead to a gigantic embedding table that causes a high memory usage cost. Second, it is likely to cause the over-fitting problem for those features that do not require too large representation capacity. Existing works that try to address the problem always cause a significant drop in recommendation performance or suff…
Department of Computer Science, City University of Hong Kong（香港城市大学计算机科学系）；Platform and Content Group, Tencent（腾讯平台与内容业务线）
Production recommendation systems rely on embedding methods to represent various features. An impeding challenge in practice is that the large embedding matrix incurs substantial memory footprint in serving as the number of features grows over time. We propose a similarity-aware embedding matrix compression method called Saec to address this challenge. Saec clusters similar features within a field to reduce the embedding matrix size. Saec also adopts a fast clustering optimization based on feature frequency to drastically improve clustering time. We implement and evaluate Saec on Numerous, the…
The gap between the randomly initialized item ID embedding and the well-trained warm item ID embedding makes the cold items hard to suit the recommendation system, which is trained on the data of historical warm items. To alleviate the performance decline of new items recommendation, the distribution of the new item ID embedding should be close to that of the historical warm items. To achieve this goal, we propose an Adversarial Variational Auto-encoder Warm-up model (AVAEW) to generate warm-up item ID embedding for cold items. Specifically, we develop a conditional variational auto-encoder mo…
Recently deep learning based recommendation systems have been actively explored to solve the cold-start problem using a hybrid approach. However, the majority of previous studies proposed a hybrid model where collaborative filtering and content-based filtering modules are independently trained. The end-to-end approach that takes different modality data as input and jointly trains the model can provide better optimization but it has not been fully explored yet. In this work, we propose deep content-user embedding model, a simple and intuitive architecture that combines the user-item interaction…
Graph Representation Learning (GRL) is an upcoming and promising area in recommendation systems. In this paper, we revisit the Singular Value Decomposition (SVD) of adjacency matrix for embedding generation of users and items and use a two-layer neural network on top of these embeddings to learn relevance between user-item pairs. Inspired by the success of higher-order learning in GRL, we further propose an extension of this method to include two-hop neighbors for SVD through the second order of the adjacency matrix and demonstrate improved performance compared with the simple SVD method which…
Nowadays designing a real recommendation system has been a critical problem for both academic and industry. However, due to the huge number of users and items, the diversity and dynamic property of the user interest, how to design a scalable recommendation system, which is able to efficiently produce effective and diverse recommendation results on billion-scale scenarios, is still a challenging and open problem for existing methods. In this paper, given the user-item interaction graph, we propose RNE, a data-efficient Recommendation-based Network Embedding method, to give personalized and dive…
Modern deep learning-based recommendation systems exploit hundreds to thousands of different categorical features, each with millions of different categories ranging from clicks to posts. To respect the natural diversity within the categorical data, embeddings map each category to a unique dense representation within an embedded space. Since each categorical feature could take on as many as tens of millions of different possible categories, the embedding tables form the primary memory bottleneck during both training and inference. We propose a novel approach for reducing the embedding size in …
Deep learning recommendation systems serve personalized content under diverse tail-latency targets and input-query loads. In order to do so, state-of-the-art recommendation models rely on terabyte-scale embedding tables to learn user preferences over large bodies of contents. The reliance on a fixed embedding representation of embedding tables not only imposes significant memory capacity and bandwidth requirements but also limits the scope of compatible system solutions. This paper challenges the assumption of fixed embedding representations by showing how synergies between embedding represent…
Deep learning-based models are utilized to achieve state-of-the-art performance for recommendation systems. A key challenge for these models is to work with millions of categorical classes or tokens. The standard approach is to learn end-to-end, dense latent representations or embeddings for each token. The resulting embeddings require large amounts of memory that blow up with the number of tokens. Training and inference with these models create storage, and memory bandwidth bottlenecks leading to significant computing and energy consumption when deployed in practice. To this end, we present t…
Graph-based recommender systems (GRSs) analyze the structural information in the graphical representation of data to make better recommendations, especially when the direct user-item relation data is sparse. Ranking-oriented GRSs that form a major class of recommendation systems, mostly use the graphical representation of preference (or rank) data for measuring node similarities, from which they can infer a recommendation list using a neighborhood-based mechanism. In this paper, we propose PGRec, a novel graph-based ranking-oriented recommendation framework. PGRec models the preferences of the…
Industry-scale recommendation systems have become a cornerstone of the e-commerce shopping experience. For Etsy, an online marketplace with over 50 million handmade and vintage items, users come to rely on personalized recommendations to surface relevant items from its massive inventory. One hallmark of Etsy's shopping experience is the multitude of ways in which a user can interact with an item they are interested in: they can view it, favorite it, add it to a collection, add it to cart, purchase it, etc. We hypothesize that the different ways in which a user interacts with an item indicates …
HPC-AI Technology Inc.（HPC-AI科技公司）；School of Computer Science, Wuhan University（武汉大学计算机学院）；National University of Singapore（新加坡国立大学）
Deep learning recommendation models (DLRMs) have been widely applied in Internet companies. The embedding tables of DLRMs are too large to fit on GPU memory entirely. We propose a GPU-based software cache approaches to dynamically manage the embedding table in the CPU and GPU memory space by leveraging the id's frequency statistics of the target dataset. Our proposed software cache is efficient in training entire DLRMs on GPU in a synchronized update manner. It is also scaled to multiple GPUs in combination with the widely used hybrid parallel training approaches. Evaluating our prototype syst…
Stanford University（斯坦福大学）；University of Washington（华盛顿大学）；University of California, Los Angeles（加州大学洛杉矶分校）；University of California, Berkeley（加州大学伯克利分校）
Recommendation systems underpin the serving of nearly all online content in the modern age. From Youtube and Netflix recommendations, to Facebook feeds and Google searches, these systems are designed to filter content to the predicted preferences of users. Recently, these systems have faced growing criticism with respect to their impact on content diversity, social polarization, and the health of public discourse. In this work we simulate the recommendations given by collaborative filtering algorithms on users in the MovieLens data set. We find that prolonged exposure to system-generated recom…
A Multimodal Single-Branch Embedding Network for Recommendation in Cold-Start and Missing Modality Scenarios
Institute of Computational Perception, Johannes Kepler University Linz（约翰开普勒林茨大学计算感知研究所）；Human-centered AI Group, AI Lab, Linz Institute of Technology（林茨理工大学人工智能实验室以人为中心的人工智能组）
Embedding representations power machine intelligence in many applications, including recommendation systems, but they are space intensive -- potentially occupying hundreds of gigabytes in large-scale settings. To help manage this outsized memory consumption, we explore mixed dimension embeddings, an embedding layer architecture in which a particular embedding vector's dimension scales with its query frequency. Through theoretical analysis and systematic experiments, we demonstrate that using mixed dimensions can drastically reduce the memory usage, while maintaining and even improving the ML p…
Most work in graph-based recommender systems considers a {\em static} setting where all information about test nodes (i.e., users and items) is available upfront at training time. However, this static setting makes little sense for many real-world applications where data comes in continuously as a stream of new edges and nodes, and one has to update model predictions incrementally to reflect the latest state. To fully capitalize on the newly available data in the stream, recent graph-based recommendation models would need to be repeatedly retrained, which is infeasible in practice. In this pap…
Latent user representations are widely adopted in the tech industry for powering personalized recommender systems. Most prior work infers a single high dimensional embedding to represent a user, which is a good starting point but falls short in delivering a full understanding of the user's interests. In this work, we introduce PinnerSage, an end-to-end recommender system that represents each user via multi-modal embeddings and leverages this rich representation of users to provides high quality personalized recommendations. PinnerSage achieves this by clustering users' actions into conceptuall…
ETH Zurich（苏黎世联邦理工学院）；Columbia University（哥伦比亚大学）；Alibaba Group（阿里巴巴集团）
Deep neural networks are widely used in personalized recommendation systems. Unlike regular DNN inference workloads, recommendation inference is memory-bound due to the many random memory accesses needed to lookup the embedding tables. The inference is also heavily constrained in terms of latency because producing a recommendation for a user must be done in about tens of milliseconds. In this paper, we propose MicroRec, a high-performance inference engine for recommendation systems. MicroRec accelerates recommendation inference by (1) redesigning the data structures involved in the embeddings …
We propose a friend recommendation system (an application of link prediction) using edge embeddings on social networks. Most real-world social networks are multi-graphs, where different kinds of relationships (e.g. chat, friendship) are possible between a pair of users. Existing network embedding techniques do not leverage signals from different edge types and thus perform inadequately on link prediction in such networks. We propose a method to mine network representation that effectively exploits heterogeneity in multi-graphs. We evaluate our model on a real-world, active social network where…
Modeling Programming Skills with Source Code Embeddings for Context-aware Exercise Recommendation
Departamento de Informática Universidade Federal de Viçosa（信息学院弗拉门多大学）；Instituto Federal de Minas Gerais（矿业联邦机构）；Instituto de Ciências Matemáticas e de Computação Universidade de São Paulo（数学与计算机科学学院圣保罗大学）；Departamento de Ciência da Computação Universidade Federal de Minas Gerais（计算机科学学院联邦矿业大学）
Personalized recommendation models (RecSys) are one of the most popular machine learning workload serviced by hyperscalers. A critical challenge of training RecSys is its high memory capacity requirements, reaching hundreds of GBs to TBs of model size. In RecSys, the so-called embedding layers account for the majority of memory usage so current systems employ a hybrid CPU-GPU design to have the large CPU memory store the memory hungry embedding layers. Unfortunately, training embeddings involve several memory bandwidth intensive operations which is at odds with the slow CPU memory, causing per…
Collecting and training over sensitive personal data raise severe privacy concerns in personalized recommendation systems, and federated learning can potentially alleviate the problem by training models over decentralized user data.However, a theoretically private solution in both the training and serving stages of federated recommendation is essential but still lacking.Furthermore, naively applying differential privacy (DP) to the two stages in federated recommendation would fail to achieve a satisfactory trade-off between privacy and utility due to the high-dimensional characteristics of mod…
Features play an important role in the prediction tasks of e-commerce recommendations. To guarantee the consistency of off-line training and on-line serving, we usually utilize the same features that are both available. However, the consistency in turn neglects some discriminative features. For example, when estimating the conversion rate (CVR), i.e., the probability that a user would purchase the item if she clicked it, features like dwell time on the item detailed page are informative. However, CVR prediction should be conducted for on-line ranking before the click happens. Thus we cannot ge…
News recommendation is important for personalized online news services. Most existing news recommendation methods rely on centrally stored user behavior data to both train models offline and provide online recommendation services. However, user data is usually highly privacy-sensitive, and centrally storing them may raise privacy concerns and risks. In this paper, we propose a unified news recommendation framework, which can utilize user data locally stored in user clients to train models and serve users in a privacy-preserving way. Following a widely used paradigm in real-world recommender sy…
University of Electronic Science and Technology of China（电子科技大学）
As a pivotal tool to alleviate the information overload problem, recommender systems aim to predict user's preferred items from millions of candidates by analyzing observed user-item relations. As for alleviating the sparsity and cold start problems encountered by recommender systems, researchers resort to employing side information or knowledge in recommendation as a strategy for uncovering hidden (indirect) user-item relations, aiming to enrich observed information (or data) for recommendation. However, in the face of the high complexity and large scale of side information and knowledge, thi…
Embedding learning of categorical features (e.g. user/item IDs) is at the core of various recommendation models including matrix factorization and neural collaborative filtering. The standard approach creates an embedding table where each row represents a dedicated embedding vector for every unique feature value. However, this method fails to efficiently handle high-cardinality features and unseen feature values (e.g. new video ID) that are prevalent in real-world recommendation systems. In this paper, we propose an alternative embedding framework Deep Hash Embedding (DHE), replacing embedding…
We introduce ZOOMER, a system deployed at Taobao, the largest e-commerce platform in China, for training and serving GNN-based recommendations over web-scale graphs. ZOOMER is designed for tackling two challenges presented by the massive user data at Taobao: low training/serving efficiency due to the huge scale of the graphs, and low recommendation quality due to the information overload which distracts the recommendation model from specific user intentions. ZOOMER achieves this by introducing a key concept, Region of Interests (ROI) in GNNs for recommendations, i.e., a neighborhood region in …
Graduate School at Shenzhen, Tsinghua University（清华大学深圳国际研究生院）；Peng Cheng Laboratory（鹏城实验室）
In the current deep learning based recommendation system, the embedding method is generally employed to complete the conversion from the high-dimensional sparse feature vector to the low-dimensional dense feature vector. However, as the dimension of the input vector of the embedding layer is too large, the addition of the embedding layer significantly slows down the convergence speed of the entire neural network, which is not acceptable in real-world scenarios. In addition, as the interaction between users and items increases and the relationship between items becomes more complicated, the emb…
Knowledge graphs (KGs) have proven to be effective for high-quality recommendation, where the connectivities between users and items provide rich and complementary information to user-item interactions. Most existing methods, however, are insufficient to exploit the KGs for capturing user preferences, as they either represent the user-item connectivities via paths with limited expressiveness or implicitly model them by propagating information over the entire KG with inevitable noise. In this paper, we design a novel hierarchical attentive knowledge graph embedding (HAKG) framework to exploit t…
To recommend relevant merchandises for seasonal retail events, we rely on item retrieval from marketplace inventory. With feedback to expand query scope, we discuss keyword expansion candidate selection using word embedding similarity, and an enhanced tf-idf formula for expanded words in search ranking.
Top-$N$ sequential recommendation models each user as a sequence of items interacted in the past and aims to predict top-$N$ ranked items that a user will likely interact in a `near future'. The order of interaction implies that sequential patterns play an important role where more recent items in a sequence have a larger impact on the next item. In this paper, we propose a Convolutional Sequence Embedding Recommendation Model (\emph{Caser}) as a solution to address this requirement. The idea is to embed a sequence of recent items into an `image' in the time and latent spaces and learn sequent…
Ping An Technology (Shenzhen) Co., Ltd（平安科技（深圳）有限公司）
The problem of session-based recommendation aims to predict user next actions based on session histories. Previous methods models session histories into sequences and estimate user latent features by RNN and GNN methods to make recommendations. However under massive-scale and complicated financial recommendation scenarios with both virtual and real commodities , such methods are not sufficient to represent accurate user latent features and neglect the long-term characteristics of users. To take long-term preference and dynamic interests into account, we propose a novel method, i.e. User-Based …
