Get our free extension to see links to code for papers anywhere online!

Chrome logo  Add to Chrome

Firefox logo Add to Firefox

"Recommendation": models, code, and papers

Live Multi-Streaming and Donation Recommendations via Coupled Donation-Response Tensor Factorization

Oct 05, 2021
Hsu-Chao Lai, Jui-Yi Tsai, Hong-Han Shuai, Jiun-Long Huang, Wang-Chien Lee, De-Nian Yang

In contrast to traditional online videos, live multi-streaming supports real-time social interactions between multiple streamers and viewers, such as donations. However, donation and multi-streaming channel recommendations are challenging due to complicated streamer and viewer relations, asymmetric communications, and the tradeoff between personal interests and group interactions. In this paper, we introduce Multi-Stream Party (MSP) and formulate a new multi-streaming recommendation problem, called Donation and MSP Recommendation (DAMRec). We propose Multi-stream Party Recommender System (MARS) to extract latent features via socio-temporal coupled donation-response tensor factorization for donation and MSP recommendations. Experimental results on Twitch and Douyu manifest that MARS significantly outperforms existing recommenders by at least 38.8% in terms of hit ratio and mean average precision.

* Proceedings of the 29th ACM International Conference on Information & Knowledge Management 1 2020 665-674 

Deep Causal Reasoning for Recommendations

Jan 06, 2022
Yaochen Zhu, Jing Yi, Jiayi Xie, Zhenzhong Chen

Traditional recommender systems aim to estimate a user's rating to an item based on observed ratings from the population. As with all observational studies, hidden confounders, which are factors that affect both item exposures and user ratings, lead to a systematic bias in the estimation. Consequently, a new trend in recommender system research is to negate the influence of confounders from a causal perspective. Observing that confounders in recommendations are usually shared among items and are therefore multi-cause confounders, we model the recommendation as a multi-cause multi-outcome (MCMO) inference problem. Specifically, to remedy confounding bias, we estimate user-specific latent variables that render the item exposures independent Bernoulli trials. The generative distribution is parameterized by a DNN with factorized logistic likelihood and the intractable posteriors are estimated by variational inference. Controlling these factors as substitute confounders, under mild assumptions, can eliminate the bias incurred by multi-cause confounders. Furthermore, we show that MCMO modeling may lead to high variance due to scarce observations associated with the high-dimensional causal space. Fortunately, we theoretically demonstrate that introducing user features as pre-treatment variables can substantially improve sample efficiency and alleviate overfitting. Empirical studies on simulated and real-world datasets show that the proposed deep causal recommender shows more robustness to unobserved confounders than state-of-the-art causal recommenders. Codes and datasets are released at


Task Recommendation in Crowdsourcing Based on Learning Preferences and Reliabilities

Jul 27, 2018
Qiyu Kang, Wee Peng Tay

Workers participating in a crowdsourcing platform can have a wide range of abilities and interests. An important problem in crowdsourcing is the task recommendation problem, in which tasks that best match a particular worker's preferences and reliabilities are recommended to that worker. A task recommendation scheme that assigns tasks more likely to be accepted by a worker who is more likely to complete it reliably results in better performance for the task requester. Without prior information about a worker, his preferences and reliabilities need to be learned over time. In this paper, we propose a multi-armed bandit (MAB) framework to learn a worker's preferences and his reliabilities for different categories of tasks. However, unlike the classical MAB problem, the reward from the worker's completion of a task is unobservable. We therefore include the use of gold tasks (i.e., tasks whose solutions are known \emph{a priori} and which do not produce any rewards) in our task recommendation procedure. Our model could be viewed as a new variant of MAB, in which the random rewards can only be observed at those time steps where gold tasks are used, and the accuracy of estimating the expected reward of recommending a task to a worker depends on the number of gold tasks used. We show that the optimal regret is $O(\sqrt{n})$, where $n$ is the number of tasks recommended to the worker. We develop three task recommendation strategies to determine the number of gold tasks for different task categories, and show that they are order optimal. Simulations verify the efficiency of our approaches.


Recommending Burgers based on Pizza Preferences: Addressing Data Sparsity with a Product of Experts

Apr 26, 2021
Martin Milenkoski, Diego Antognini, Claudiu Musat

In this paper we describe a method to tackle data sparsity and create recommendations in domains with limited knowledge about the user preferences. We expand the variational autoencoder collaborative filtering from a single-domain to a multi domain setting. The intuition is that user-item interactions in a source domain can augment the recommendation quality in a target domain. The intuition can be taken to its extreme, where, in a cross-domain setup, the user history in a source domain is enough to generate high quality recommendations in a target one. We thus create a Product-of-Experts (POE) architecture for recommendations that jointly models user-item interactions across multiple domains. The method is resilient to missing data for one or more of the domains, which is a situation often found in real life. We present results on two widely-used datasets - Amazon and Yelp, which support the claim that holistic user preference knowledge leads to better recommendations. Surprisingly, we find that in select cases, a POE recommender that does not access the target domain user representation can surpass a strong VAE recommender baseline trained on the target domain. We complete the analysis with a study of the reasons behind this outperformance and an in-depth look at the resulting embedding spaces.

* Under review. 16 pages, 5 figures, 2 tables 

A Soft Recommender System for Social Networks

Jan 08, 2020
Marzieh Pourhojjati-Sabet, Azam Rabiee

Recent social recommender systems benefit from friendship graph to make an accurate recommendation, believing that friends in a social network have exactly the same interests and preferences. Some studies have benefited from hard clustering algorithms (such as K-means) to determine the similarity between users and consequently to define degree of friendships. In this paper, we went a step further to identify true friends for making even more realistic recommendations. we calculated the similarity between users, as well as the dependency between a user and an item. Our hypothesis is that due to the uncertainties in user preferences, the fuzzy clustering, instead of the classical hard clustering, is beneficial in accurate recommendations. We incorporated the C-means algorithm to get different membership degrees of soft users' clusters. Then, the users' similarity metric is defined according to the soft clusters. Later, in a training scheme we determined the latent representations of users and items, extracting from the huge and sparse user-item-tag matrix using matrix factorization. In the parameter tuning, we found the optimum coefficients for the influence of our soft social regularization and the user-item dependency terms. Our experimental results convinced that the proposed fuzzy similarity metric improves the recommendations in real data compared to the baseline social recommender system with the hard clustering.

* 8 pages, 6 figures 

DeSkew-LSH based Code-to-Code Recommendation Engine

Nov 05, 2021
Fran Silavong, Sean Moran, Antonios Georgiadis, Rohan Saphal, Robert Otter

Machine learning on source code (MLOnCode) is a popular research field that has been driven by the availability of large-scale code repositories and the development of powerful probabilistic and deep learning models for mining source code. Code-to-code recommendation is a task in MLOnCode that aims to recommend relevant, diverse and concise code snippets that usefully extend the code currently being written by a developer in their development environment (IDE). Code-to-code recommendation engines hold the promise of increasing developer productivity by reducing context switching from the IDE and increasing code-reuse. Existing code-to-code recommendation engines do not scale gracefully to large codebases, exhibiting a linear growth in query time as the code repository increases in size. In addition, existing code-to-code recommendation engines fail to account for the global statistics of code repositories in the ranking function, such as the distribution of code snippet lengths, leading to sub-optimal retrieval results. We address both of these weaknesses with \emph{Senatus}, a new code-to-code recommendation engine. At the core of Senatus is \emph{De-Skew} LSH a new locality sensitive hashing (LSH) algorithm that indexes the data for fast (sub-linear time) retrieval while also counteracting the skewness in the snippet length distribution using novel abstract syntax tree-based feature scoring and selection algorithms. We evaluate Senatus via automatic evaluation and with an expert developer user study and find the recommendations to be of higher quality than competing baselines, while achieving faster search. For example, on the CodeSearchNet dataset we show that Senatus improves performance by 6.7\% F1 and query time 16x is faster compared to Facebook Aroma on the task of code-to-code recommendation.


Towards Topic-Guided Conversational Recommender System

Oct 08, 2020
Kun Zhou, Yuanhang Zhou, Wayne Xin Zhao, Xiaoke Wang, Ji-Rong Wen

Conversational recommender systems (CRS) aim to recommend high-quality items to users through interactive conversations. To develop an effective CRS, the support of high-quality datasets is essential. Existing CRS datasets mainly focus on immediate requests from users, while lack proactive guidance to the recommendation scenario. In this paper, we contribute a new CRS dataset named \textbf{TG-ReDial} (\textbf{Re}commendation through \textbf{T}opic-\textbf{G}uided \textbf{Dial}og). Our dataset has two major features. First, it incorporates topic threads to enforce natural semantic transitions towards the recommendation scenario. Second, it is created in a semi-automatic way, hence human annotation is more reasonable and controllable. Based on TG-ReDial, we present the task of topic-guided conversational recommendation, and propose an effective approach to this task. Extensive experiments have demonstrated the effectiveness of our approach on three sub-tasks, namely topic prediction, item recommendation and response generation. TG-ReDial is available at

* 12 pages, Accepted by Coling2020 

Regret in Online Recommendation Systems

Oct 23, 2020
Kaito Ariu, Narae Ryu, Se-Young Yun, Alexandre Proutière

This paper proposes a theoretical analysis of recommendation systems in an online setting, where items are sequentially recommended to users over time. In each round, a user, randomly picked from a population of $m$ users, requests a recommendation. The decision-maker observes the user and selects an item from a catalogue of $n$ items. Importantly, an item cannot be recommended twice to the same user. The probabilities that a user likes each item are unknown. The performance of the recommendation algorithm is captured through its regret, considering as a reference an Oracle algorithm aware of these probabilities. We investigate various structural assumptions on these probabilities: we derive for each structure regret lower bounds, and devise algorithms achieving these limits. Interestingly, our analysis reveals the relative weights of the different components of regret: the component due to the constraint of not presenting the same item twice to the same user, that due to learning the chances users like items, and finally that arising when learning the underlying structure.

* Advances in Neural Information Processing Systems (NeurIPS 2020) 

Explicit User Manipulation in Reinforcement Learning Based Recommender Systems

Mar 20, 2022
Matthew Sparr

Recommender systems are highly prevalent in the modern world due to their value to both users and platforms and services that employ them. Generally, they can improve the user experience and help to increase satisfaction, but they do not come without risks. One such risk is that of their effect on users and their ability to play an active role in shaping user preferences. This risk is more significant for reinforcement learning based recommender systems. These are capable of learning for instance, how recommended content shown to a user today may tamper that user's preference for other content recommended in the future. Reinforcement learning based recommendation systems can thus implicitly learn to influence users if that means maximizing clicks, engagement, or consumption. On social news and media platforms, in particular, this type of behavior is cause for alarm. Social media undoubtedly plays a role in public opinion and has been shown to be a contributing factor to increased political polarization. Recommender systems on such platforms, therefore, have great potential to influence users in undesirable ways. However, it may also be possible for this form of manipulation to be used intentionally. With advancements in political opinion dynamics modeling and larger collections of user data, explicit user manipulation in which the beliefs and opinions of users are tailored towards a certain end emerges as a significant concern in reinforcement learning based recommender systems.


Unintended Bias in Language Model-driven Conversational Recommendation

Jan 19, 2022
Tianshu Shen, Jiaru Li, Mohamed Reda Bouadjenek, Zheda Mai, Scott Sanner

Conversational Recommendation Systems (CRSs) have recently started to leverage pretrained language models (LM) such as BERT for their ability to semantically interpret a wide range of preference statement variations. However, pretrained LMs are well-known to be prone to intrinsic biases in their training data, which may be exacerbated by biases embedded in domain-specific language data(e.g., user reviews) used to fine-tune LMs for CRSs. We study a recently introduced LM-driven recommendation backbone (termed LMRec) of a CRS to investigate how unintended bias i.e., language variations such as name references or indirect indicators of sexual orientation or location that should not affect recommendations manifests in significantly shifted price and category distributions of restaurant recommendations. The alarming results we observe strongly indicate that LMRec has learned to reinforce harmful stereotypes through its recommendations. For example, offhand mention of names associated with the black community significantly lowers the price distribution of recommended restaurants, while offhand mentions of common male-associated names lead to an increase in recommended alcohol-serving establishments. These and many related results presented in this work raise a red flag that advances in the language handling capability of LM-drivenCRSs do not come without significant challenges related to mitigating unintended bias in future deployed CRS assistants with a potential reach of hundreds of millions of end-users.

* 12 pages, 7 figures