Recommender systems operate in an inherently dynamical setting. Past recommendations influence future behavior, including which data points are observed and how user preferences change. However, experimenting in production systems with real user dynamics is often infeasible, and existing simulation-based approaches have limited scale. As a result, many state-of-the-art algorithms are designed to solve supervised learning problems, and progress is judged only by offline metrics. In this work we investigate the extent to which offline metrics predict online performance by evaluating eleven recommenders across six controlled simulated environments. We observe that offline metrics are correlated with online performance over a range of environments. However, improvements in offline metrics lead to diminishing returns in online performance. Furthermore, we observe that the ranking of recommenders varies depending on the amount of initial offline data available. We study the impact of adding exploration strategies, and observe that their effectiveness, when compared to greedy recommendation, is highly dependent on the recommendation algorithm. We provide the environments and recommenders described in this paper as Reclab: an extensible ready-to-use simulation framework at https://github.com/berkeley-reclab/RecLab.
In academic literature, recommender systems are often evaluated on the task of next-item prediction. The procedure aims to give an answer to the question: "Given the natural sequence of user-item interactions up to time t, can we predict which item the user will interact with at time t+1?". Evaluation results obtained through said methodology are then used as a proxy to predict which system will perform better in an online setting. The online setting, however, poses a subtly different question: "Given the natural sequence of user-item interactions up to time t, can we get the user to interact …
Both in academic and industry-based research, online evaluation methods are seen as the golden standard for interactive applications like recommendation systems. Naturally, the reason for this is that we can directly measure utility metrics that rely on interventions, being the recommendations that are being shown to users. Nevertheless, online evaluation methods are costly for a number of reasons, and a clear need remains for reliable offline evaluation procedures. In industry, offline metrics are often used as a first-line evaluation to generate promising candidate models to evaluate online.…
Moscow Institute of Physics and Technology（莫斯科物理技术学院）
We explore a hidden feedback loops effect in online recommender systems. Feedback loops result in degradation of online multi-armed bandit (MAB) recommendations to a small subset and loss of coverage and novelty. We study how uncertainty and noise in user interests influence the existence of feedback loops. First, we show that an unbiased additive random noise in user interests does not prevent a feedback loop. Second, we demonstrate that a non-zero probability of resetting user interests is sufficient to limit the feedback loop and estimate the size of the effect. Our experiments confirm the …
Before A/B testing online a new version of a recommender system, it is usual to perform some offline evaluations on historical data. We focus on evaluation methods that compute an estimator of the potential uplift in revenue that could generate this new technology. It helps to iterate faster and to avoid losing money by detecting poor policies. These estimators are known as counterfactual or off-policy estimators. We show that traditional counterfactual estimators such as capped importance sampling and normalised importance sampling are experimentally not having satisfying bias-variance compro…
Many video-on-demand and music streaming services provide the user with a page consisting of several recommendation lists, i.e. widgets or swipeable carousels, each built with a specific criterion (e.g. most recent, TV series, etc.). Finding efficient strategies to select which carousels to display is an active research topic of great industrial interest. In this setting, the overall quality of the recommendations of a new algorithm cannot be assessed by measuring solely its individual recommendation quality. Rather, it should be evaluated in a context where other recommendation lists are alre…
Swisscom（瑞士电信）；Ecole Polytechnique Fédérale de Lausanne（洛桑联邦理工学院）；Symphony
Recommender systems research tends to evaluate model performance offline and on randomly sampled targets, yet the same systems are later used to predict user behavior sequentially from a fixed point in time. Simulating online recommender system performance is notoriously difficult and the discrepancy between online and offline behaviors is typically not accounted for in offline evaluations. This disparity permits weaknesses to go unnoticed until the model is deployed in a production setting. In this paper, we first demonstrate how omitting temporal context when evaluating recommender system pe…
ZoRRO: A Zero-Weight Personalized Recommender System for Scalable News Recommendation
Technical University of Denmark（丹麦技术大学）；ZOZO Research（ZOZO研究公司）；University of California San Diego（加州大学圣地亚哥分校）；Pioneer Centre for Artificial Intelligence（先锋人工智能中心）
Advertising expenditures have become the major source of revenue for e-commerce platforms. Providing good advertising experiences for advertisers by reducing their costs of trial and error in discovering the optimal advertising strategies is crucial for the long-term prosperity of online advertising. To achieve this goal, the advertising platform needs to identify the advertiser's optimization objectives, and then recommend the corresponding strategies to fulfill the objectives. In this work, we first deploy a prototype of strategy recommender system on Taobao display advertising platform, whi…
Time to Split: Exploring Data Splitting Strategies for Offline Evaluation of Sequential Recommenders
AIRI, Skoltech Moscow Russian Federation（AIRI，斯克里普钦斯基因工学院莫斯科俄罗斯联邦）；Sber AI Lab, Skoltech Moscow Russian Federation（Sber AI Lab，斯克里普钦斯基因工学院莫斯科俄罗斯联邦）；Sber AI Lab Moscow Russian Federation（Sber AI Lab，莫斯科斯克里普钦斯基因工学院俄罗斯联邦）；Sber AI Lab, HSE University Moscow Russian Federation（Sber AI Lab，莫斯科高等经济大学俄罗斯联邦）
Recurrent recommender systems have been successful in capturing the temporal dynamics in users' activity trajectories. However, recurrent neural networks (RNNs) are known to have difficulty learning long-term dependencies. As a consequence, RNN-based recommender systems tend to overly focus on short-term user interests. This is referred to as the recency bias, which could negatively affect the long-term user experience as well as the health of the ecosystem. In this paper, we introduce the recency dropout technique, a simple yet effective data augmentation technique to alleviate the recency bi…
HC Research（HC研究院）；HC Financial Service Group（HC金融服务集团）
Recommender systems have received great commercial success. Recommendation has been used widely in areas such as e-commerce, online music FM, online news portal, etc. However, several problems related to input data structure pose serious challenge to recommender system performance. Two of these problems are Matthew effect and sparsity problem. Matthew effect heavily skews recommender system output towards popular items. Data sparsity problem directly affects the coverage of recommendation result. Collaborative filtering is a simple benchmark ubiquitously adopted in the industry as the baseline…
Modern online services continuously generate data at very fast rates. This continuous flow of data encompasses content - e.g., posts, news, products, comments -, but also user feedback - e.g., ratings, views, reads, clicks -, together with context data - user device, spatial or temporal data, user task or activity, weather. This can be overwhelming for systems and algorithms designed to train in batches, given the continuous and potentially fast change of content, context and user preferences or intents. Therefore, it is important to investigate online methods able to transparently adapt to th…
University of Klagenfurt（克拉根福大学）；TU Dortmund（多特蒙德工业大学）
Recommender Systems are nowadays successfully used by all major web sites (from e-commerce to social media) to filter content and make suggestions in a personalized way. Academic research largely focuses on the value of recommenders for consumers, e.g., in terms of reduced information overload. To what extent and in which ways recommender systems create business value is, however, much less clear, and the literature on the topic is scattered. In this research commentary, we review existing publications on field tests of recommender systems and report which business-related performance measures…
Intelligent personal assistants (IPA) enable voice applications that facilitate people's daily tasks. However, due to the complexity and ambiguity of voice requests, some requests may not be handled properly by the standard natural language understanding (NLU) component. In such cases, a simple reply like "Sorry, I don't know" hurts the user's experience and limits the functionality of IPA. In this paper, we propose a two-stage shortlister-reranker recommender system to match third-party voice applications (skills) to unhandled utterances. In this approach, a skill shortlister is proposed to r…
Recall assistance methods are among the key aspects that improve the accuracy of online dietary assessment surveys. These methods still mainly rely on experience of trained interviewers with nutritional background, but data driven approaches could improve cost-efficiency and scalability of automated dietary assessment. We evaluated the effectiveness of a recommender algorithm developed for an online dietary assessment system called Intake24, that automates the multiple-pass 24-hour recall method. The recommender builds a model of eating behavior from recalls collected in past surveys. Based on…
Recommender systems are facing scrutiny because of their growing impact on the opportunities we have access to. Current audits for fairness are limited to coarse-grained parity assessments at the level of sensitive groups. We propose to audit for envy-freeness, a more granular criterion aligned with individual preferences: every user should prefer their recommendations to those of other users. Since auditing for envy requires to estimate the preferences of users beyond their existing recommendations, we cast the audit as a new pure exploration problem in multi-armed bandits. We propose a sampl…
In this paper, we consider decentralized sequential decision making in distributed online recommender systems, where items are recommended to users based on their search query as well as their specific background including history of bought items, gender and age, all of which comprise the context information of the user. In contrast to centralized recommender systems, in which there is a single centralized seller who has access to the complete inventory of items as well as the complete record of sales and user information, in decentralized recommender systems each seller/learner only has acces…
We study offline recommender learning from explicit rating feedback in the presence of selection bias. A current promising solution for the bias is the inverse propensity score (IPS) estimation. However, the performance of existing propensity-based methods can suffer significantly from the propensity estimation bias. In fact, most of the previous IPS-based methods require some amount of missing-completely-at-random (MCAR) data to accurately estimate the propensity. This leads to a critical self-contradiction; IPS is ineffective without MCAR data, even though it originally aims to learn recomme…
Different software tools have been developed with the purpose of performing offline evaluations of recommender systems. However, the results obtained with these tools may be not directly comparable because of subtle differences in the experimental protocols and metrics. Furthermore, it is difficult to analyze in the same experimental conditions several algorithms without disclosing their implementation details. For these reasons, we introduce RecLab, an open source software for evaluating recommender systems in a distributed fashion. By relying on consolidated web protocols, we created RESTful…
Multi-task Offline Reinforcement Learning for Online Advertising in Recommender Systems
Taobao \& Tmall Group of Alibaba City University of Hong Kong Hangzhou China；Southern University of Science；Technology, City University of Hong Kong Hong Kong China；City University of Hong Kong Hong Kong China；University of Queensland Brisbane Australia；Taobao \& Tmall Group of Alibaba Beijing China；Taobao \& Tmall Group of Alibaba City University of Hong Kong；Technology, City University of Hong Kong；City University of Hong Kong；University of Queensland；Taobao \& Tmall Group of Alibaba
In this paper, we argue that the paradigm commonly adopted for offline evaluation of sequential recommender systems is unsuitable for evaluating reinforcement learning-based recommenders. We find that most of the existing offline evaluation practices for reinforcement learning-based recommendation are based on a next-item prediction protocol, and detail three shortcomings of such an evaluation protocol. Notably, it cannot reflect the potential benefits that reinforcement learning (RL) is expected to bring while it hides critical deficiencies of certain offline RL agents. Our suggestions for al…
Online dating platforms provide people with the opportunity to find a partner. Recommender systems in online dating platforms suggest one side of users to the other side of users. We discuss the potential interactions between reciprocal recommender systems (RRSs) and matching theory. We present our ongoing project to deploy a matching theory-based recommender system (MTRS) in a real-world online dating platform.
Various studies in recent years have pointed out large issues in the offline evaluation of recommender systems, making it difficult to assess whether true progress has been made. However, there has been little research into what set of practices should serve as a starting point during experimentation. In this paper, we examine four larger issues in recommender system research regarding uncertainty estimation, generalization, hyperparameter optimization and dataset pre-processing in more detail to arrive at a set of guidelines. We present a TrainRec, a lightweight and flexible toolkit for offli…
Recommendations are broadly used in marketplaces to match users with items relevant to their interests and needs. To understand user intent and tailor recommendations to their needs, we use deep learning to explore various heterogeneous data available in marketplaces. This paper focuses on the challenge of measuring recommender performance and summarizes the online experiment results with several promising types of deep neural network recommenders - hybrid item representation models combining features from user engagement and content, sequence-based models, and multi-armed bandit models that o…
Online rating platform represents the new trend of online cultural and commercial goods consumption. The user rating data on such platforms are foods for recommender system algorithms. Understanding the evolution pattern and its underlying mechanism is the key to understand the structures of input data for recommender systems. Prior research on input data analysis for recommender systems is quite limited, with a notable exception in 2018 [6]. In this paper, we take advantage of Poisson Process to analyze the evolution mechanism of the input data structures. We discover that homogeneous Poisson…
Online recommender systems often deal with continuous, potentially fast and unbounded flows of data. Ensemble methods for recommender systems have been used in the past in batch algorithms, however they have never been studied with incremental algorithms that learn from data streams. We evaluate online bagging with an incremental matrix factorization algorithm for top-N recommendation with positive-only -- binary -- ratings. Our results show that online bagging is able to improve accuracy up to 35% over the baseline, with small computational overhead.
Recommendation systems are a vital component of many online marketplaces, where there are often millions of items to potentially present to users who have a wide variety of wants or needs. Evaluating recommender system algorithms is a hard task, given all the inherent bias in the data, and successful companies must be able to rapidly iterate on their solution to maintain their competitive advantage. The gold standard for evaluating recommendation algorithms has been the A/B test since it is an unbiased way to estimate how well one or more algorithms compare in the real world. However, there ar…
In this paper, we present work-in-progress on SocRecM, a novel social recommendation framework for online marketplaces. We demonstrate that SocRecM is not only easy to integrate with existing Web technologies through a RESTful, scalable and easy-to-extend service-based architecture but also reveal the extent to which various social features and recommendation approaches are useful in an online social marketplace environment.
Real-world item recommenders commonly suffer from a persistent cold-start problem which is caused by dynamically changing users and items. In order to overcome the problem, several context-aware recommendation techniques have been recently proposed. In terms of both feasibility and performance, factorization machine (FM) is one of the most promising methods as generalization of the conventional matrix factorization techniques. However, since online algorithms are suitable for dynamic data, the static FMs are still inadequate. Thus, this paper proposes incremental FMs (iFMs), a general online f…
Offline evaluations of recommender systems attempt to estimate users' satisfaction with recommendations using static data from prior user interactions. These evaluations provide researchers and developers with first approximations of the likely performance of a new system and help weed out bad ideas before presenting them to users. However, offline evaluation cannot accurately assess novel, relevant recommendations, because the most novel items were previously unknown to the user, so they are missing from the historical data and cannot be judged as relevant. We present a simulation study to es…
Users of online dating sites are facing information overload that requires them to manually construct queries and browse huge amount of matching user profiles. This becomes even more problematic for multimedia profiles. Although matchmaking is frequently cited as a typical application for recommender systems, there is a surprising lack of work published in this area. In this paper we describe a recommender system we implemented and perform a quantitative comparison of two collaborative filtering (CF) and two global algorithms. Results show that collaborative filtering recommenders significantl…
