- PABRUCE – Bundle Recommendation Using Contextualized item Embeddings
by Tzoof Avny Brosh (Ben Gurion , Israel), Amit Livne (Ben-Gurion University of the Negev, Israel, Ben-Gurion University of the Negev, Israel), Oren Sar Shalom (Facebook, Israel, Facebook, Israel), Bracha Shapira (Ben-Gurion University of the Negev, Israel, Ben-Gurion University of the Negev, Israel), Mark Last (Ben-Gurion University of the Negev, Israel, Ben-Gurion University of the Negev, Israel)
A bundle is a pre-defined set of items that are collected together. In many domains, bundling is one of the most important marketing strategies for item promotion, commonly used in e-commerce. Bundle recommendation resembles the item recommendation task, where bundles are the recommended unit, but it poses additional challenges; while item recommendation requires only user and item understanding, bundle recommendation also requires modeling the connections between the various items in a bundle.
Transformers have driven the state-of-the-art methods for set and sequence modeling in various natural language processing and computer vision tasks, emphasizing the understanding that the neighbors of an element are of crucial importance. Under some required adjustments, we believe the same applies for items in bundles, and better capturing the relations of an item with other items in the bundle may
lead to improved recommendations.
To address that, we introduce BRUCE – a novel model for bundle recommendation, in which we adapt Transformers to represent data on users, items, and bundles. This allows exploiting the self-attention mechanism to model the following: latent relations between the items in a bundle; and users’ preferences toward each of the items in the bundle and toward the whole bundle.
Moreover, we examine various architectures to integrate the items’ and the users’ information and provide insights on architecture selection based on data characteristics.
Experiments conducted on three benchmark datasets show that the proposed approach contributes to the accuracy of the recommendation and substantially outperforms state-of-the-art methods.
Full text in ACM Digital Library
|
- PAProtoMF: Prototype-based Matrix Factorization for Effective and Explainable Recommendations
by Alessandro B. Melchiorre (Johannes Kepler University, Austria, Human-centered AI Group, AI Lab, Linz Institute of Technology, Austria), Navid Rekabsaz (Johannes Kepler University, Austria, Human-centered AI Group, AI Lab, Linz Institute of Technology, Austria), Christian Ganhör (Johannes Kepler University, Austria), Markus Schedl (Johannes Kepler University Linz, Austria, Human-centered AI Group, AI Lab, Linz Institute of Technology, Austria)
Recent studies show the benefits of reformulating common machine learning models through the concept of prototypes — representatives of the underlying data, used to calculate the prediction score as a linear combination of similarities of a data point to prototypes. Such prototype-based formulation of a model, in addition to preserving (sometimes enhancing) the performance, enables explainability of the model’s decisions, as the prediction can be linearly broken down into the contributions of distinct definable prototypes. Following this direction, we extend the idea of prototypes to the recommender system domain by introducing ProtoMF, a novel collaborative filtering algorithm. ProtoMF first learns sets of user/item prototypes, representing the general consumption characteristics of users/items in the underlying dataset. Using these prototypes, ProtoMF then represents users and items as vectors of similarities to the corresponding prototypes. These user/item representations are ultimately leveraged to make recommendations that are both effective in terms of accuracy metrics, and explainable through the interpretation of prototypes’ contributions to the affinity scores. We conduct experiments on three datasets to assess both the effectiveness and the explainability of ProtoMF. Addressing the former, we show that ProtoMF exhibits higher Hit~Ratio and NDCG compared to other relevant collaborative filtering approaches. As for the latter, we qualitatively show how ProtoMF can provide explainable recommendations and how its explanation capabilities can expose the existence of statistical biases in the learned representations, which we exemplify for the case of gender bias.
Full text in ACM Digital Library
|
- PATinyKG: Memory-Efficient Training Framework for Knowledge Graph Neural Recommender Systems
by Huiyuan Chen (Visa Research, United States, Visa Research, United States), Xiaoting Li (Visa Research , United States, Visa Research , United States), Kaixiong Zhou (Rice University, United States), Xia Hu (Rice University, United States), Chin-Chia Michael Yeh (Visa Inc, United States, Visa Inc, United States), Yan Zheng (Visa Research, United States, Visa Research, United States), Hao Yang (Visa Research, United States, Visa Research, United States)
Training Knowledge Graph Neural Networks (KGNNs) on large graphs is a fundamental challenge due to the high memory usage, which is mainly occupied by activations
(e.g., node embeddings). Previous works usually focus on reducing the number of nodes retained in memory. In parallel, unlike what has been developed
for other types of neural networks, training with compressed activation maps is
less explored for KGNNs. This extension is notoriously difficult to implement
due to the lack of necessary tools in common graph learning packages. To unleash the potential of this direction, we provide an optimized GPU implementation which supports training GNNs with compressed activations. In this work, we try to propose a memory-efficient knowledge graph completion via compressing
activation maps for recommendations, called TinyKG.
Full text in ACM Digital Library
|
- PAGlobal and Personalized Graphs for Heterogeneous Sequential Recommendation by Learning Behavior Transitions and User Intentions
by Weixin Chen (Shenzhen University, China), Mingkai He (Shenzhen University, China), Yongxin Ni (National University of Singapore, Singapore), Weike Pan (Shenzhen University, China), Li Chen (Hong Kong Baptist University, Hong Kong), Zhong Ming (Shenzhen University, China)
Heterogeneous sequential recommendation (HSR) is a very important recommendation problem, which aims to predict a user’s next interacted item under a target behavior type (e.g., purchase in e-commerce sites) based on his/her historical interactions with different behaviors. Though existing sequential methods have achieved advanced performance by considering the varied impacts of interactions with sequential information, a large body of them still have two major shortcomings. Firstly, they model different behaviors separately without considering the correlations between them. The transitions from item to item under diverse behaviors indicate some users’ potential behavior manner. Secondly, though the behavior information contains a user’s fine-grained interests, the insufficient consideration of the local context information limits them from well understanding user intentions. Utilizing the adjacent interactions to better understand a user’s behavior could improve the certainty of prediction. To address these two issues, we propose a novel solution utilizing global and personalized graphs for HSR (GPG4HSR) to learn behavior transitions and user intentions. Specifically, our GPG4HSR consists of two graphs, i.e., a global graph to capture the transitions between different behaviors, and a personalized graph to model items with behaviors by further considering the distinct user intentions of the adjacent contextually relevant nodes. Extensive experiments on four public datasets with the state-of-the-art baselines demonstrate the effectiveness and general applicability of our method GPG4HSR.
Full text in ACM Digital Library
|
- PACAEN: A Hierarchically Attentive Evolution Network for Item-Attribute-Change-Aware Recommendation in the Growing E-commerce Environment
by Rui Ma (Alibaba Group, China), Ning Liu (Tsinghua University, China), Jingsong Yuan (Alibaba Group, China), Huafeng Yang (Alibaba Group, China), Jiandong Zhang (Alibaba Group, China)
Traditional recommendation systems mainly focus on modeling user interests. However, the dynamics of recommended items caused by attribute modifications (e.g. changes in product prices) are also of great importance in real systems, especially under the fast-growing e-commerce environment, which may cause the users’ demands to emerge, shift and disappear. Recent studies that make efforts on dynamic item representations treat the item attributes as side information but ignore its temporal dependency, or model the item evolution with a sequence of related users but do not consider item attributes. In this paper, we propose Core Attribute Evolution Network (CAEN), which partitions the user sequence according to the attribute value and thus models the item evolution over attribute dynamics with these users. Under this framework, we further devise a hierarchical attention mechanism that applies an attribute-aware attention for user aggregation under each attribute, as well as a personalized attention for activating similar users in assessing the matching degree between target user and item. Results from the extensive experiments over actual e-commerce datasets show that our approach outperforms the state-of-art methods and achieves significant improvements on the items with rapid changes over attributes, therefore help the item recommendation to adapt to the growth of e-commerce platform.
Full text in ACM Digital Library
|
- INTorchRec: a PyTorch domain library for recommendation systems
by Dmytro Ivchenko (Meta AI, United States), Dennis Van Der Staay (Meta AI, United States), Colin Taylor (Meta AI, United States), Xing Liu (Meta AI, United States), Will Feng (Meta AI, United States), Rahul Kindi (Meta AI, United States), Anirudh Sudarshan (Meta AI, United States), Shahin Sefati (Meta AI, United States)
Recommendation Systems (RecSys) comprise a large footprint of production-deployed AI today. The neural network-based recommender systems differ from deep learning models in other domains in using high-cardinality categorical sparse features that require large embedding tables to be trained. In this talk we introduce TorchRec, a PyTorch domain library for Recommendation Systems. This new library provides common sparsity and parallelism primitives, enabling researchers to build state-of-the-art personalization models and deploy them in production. In this talk we cover the building blocks of the TorchRec library including modeling primitives such as embedding bags and jagged tensors, optimized recommender system kernels powered by FBGEMM, a flexible sharder that supports a veriety of strategies for partitioning embedding tables, a planner that automatically generates optimized and performant sharding plans, support for GPU inference and common modeling modules for building recommender system models. TorchRec library is currently used to train most large-scale recommender models at Meta. We will present how TorchRec enabled Meta’s recommender system platform to transition from CPU asynchronous training to accelerator-based full-sync training.
Full text in ACM Digital Library
|
- PABundle MCR: Towards Conversational Bundle Recommendation
by Zhankui He (UC San Diego, United States, UC San Diego, United States), Handong Zhao (Adobe Research, United States, Adobe Research, United States), Tong Yu (Adobe Research, United States, Adobe Research, United States), Sungchul Kim (Adobe Research, United States), Fan Du (Adobe Research, United States, Adobe Research, United States), Julian McAuley (UC San Diego, United States, UC San Diego, United States)
Bundle recommender systems recommend sets of items (e.g., pants, shirt, and shoes) to users, but they often suffer from two issues: significant interaction sparsity and a large output space. In this work, we extend multi-round conversational recommendation (MCR) to alleviate these issues. MCR—which uses a conversational paradigm to elicit user interests by asking user preferences on tags (e.g., categories or attributes) and handling user feedback across multiple rounds—is an emerging recommendation setting to acquire user feedback and narrow down the output space, but has not been explored in the context of bundle recommendation.
In this work, we propose a novel recommendation task named Bundle MCR. Unlike traditional bundle recommendation (considering of e.g. a bundle-aware user model and bundle generation), Bundle MCR studies how to encode user feedback as conversation states and how to post questions to users. Unlike existing MCR in which agents recommend individual items only, Bundle MCR handles more complicated user feedback on multiple items and related tags. To support this, we first propose a new framework to formulate Bundle MCR as Markov Decision Processes (MDPs) with multiple agents, for user modeling, consultation and feedback handling in bundle contexts. Under this framework, we propose a model architecture, called Bundle Bert (Bunt) to (1) recommend items, (2) post questions and (3) manage conversations based on bundle-aware conversation states. Moreover, to train Bunt effectively, we propose a two-stage training strategy. In an offline pre-training stage, Bunt is trained using multiple cloze tasks to mimic bundle interactions in conversations. Then in an online fine-tuning stage, Bunt agents are enhanced by user interactions. Our experiments on multiple offline datasets as well as the human evaluation show the value of extending MCR frameworks to bundle settings and the effectiveness of our Bunt design.
Full text in ACM Digital Library
|
- PARecommendation as Language Processing (RLP): A Unified Pretrain, Personalized Prompt & Predict Paradigm (P5)
by Shijie Geng (Rutgers University, United States), Shuchang Liu (Rutgers University, United States), Zuohui Fu (Rutgers University, United States), Yingqiang Ge (Rutgers University, United States), Yongfeng Zhang (Rutgers University, United States)
For a long period, different recommendation tasks typically require designing task-specific architectures and training objectives. As a result, it is hard to transfer the learned knowledge and representations from one task to another, thus restricting the generalization ability of existing recommendation approaches, e.g., a sequential recommendation model can hardly be applied or transferred to a review generation method. To deal with such issues, considering that language grounding is a powerful medium to describe and represent various problems or tasks, we present a flexible and unified text-to-text paradigm called “Pretrain, Personalized Prompt, and Predict Paradigm” (P5) for recommendation, which unifies various recommendation tasks in a shared framework. In P5, all data such as user-item interactions, item metadata, and user reviews are converted to a common format — natural language sequences. The rich information from natural language assists P5 to capture deeper semantics for recommendation. Specifically, P5 learns different tasks with the same language modeling objective during pretraining. Thus, it possesses the potential to serve as the foundation model for downstream recommendation tasks, allows easy integration with other modalities, and enables a unified recommendation engine. With adaptive personalized prompt for different users, P5 is able to make predictions in a zero-shot or few-shot manner and largely reduces the necessity for extensive fine-tuning. On several recommendation benchmarks, we conduct experiments to show the effectiveness of our generative approach. We will release our prompts and pretrained P5 language model to help advance future research on Recommendation as Language Processing (RLP) and Personalized Foundation Models.
Full text in ACM Digital Library
|
