All learning algorithms for recommendations face inevitable and critical trade-off between exploiting partial knowledge of a user's preferences for short-term satisfaction and exploring additional user preferences for long-term coverage. Although exploration is indispensable for long success of a recommender system, the exploration has been considered as the risk to decrease user satisfaction. The reason for the risk is that items chosen for exploration frequently mismatch with the user's interests. To mitigate this risk, recommender systems have mixed items chosen for exploration into a recommendation list, disguising the items as recommendations to elicit feedback on the items to discover the user's additional tastes. This mix-in approach has been widely used in many recommenders, but there is rare research, evaluating the effectiveness of the mix-in approach or proposing a new approach for eliciting user feedback without deceiving users. In this work, we aim to propose a new approach for feedback elicitation without any deception and compare our approach to the conventional mix-in approach for evaluation. To this end, we designed a recommender interface that reveals which items are for exploration and conducted a within-subject study with 94 MTurk workers. Our results indicated that users left significantly more feedback on items chosen for exploration with our interface. Besides, users evaluated that our new interface is better than the conventional mix-in interface in terms of novelty, diversity, transparency, trust, and satisfaction. Finally, path analysis show that, in only our new interface, exploration caused to increase user-centric evaluation metrics. Our work paves the way for how to design an interface, which utilizes learning algorithm based on users' feedback signals, giving better user experience and gathering more feedback data.
In this paper, we provide a bilingual parallel human-to-human recommendation dialog dataset (DuRecDial 2.0) to enable researchers to explore a challenging task of multilingual and cross-lingual conversational recommendation. The difference between DuRecDial 2.0 and existing conversational recommendation datasets is that the data item (Profile, Goal, Knowledge, Context, Response) in DuRecDial 2.0 is annotated in two languages, both English and Chinese, while other datasets are built with the setting of a single language. We collect 8.2k dialogs aligned across English and Chinese languages (16.5k dialogs and 255k utterances in total) that are annotated by crowdsourced workers with strict quality control procedure. We then build monolingual, multilingual, and cross-lingual conversational recommendation baselines on DuRecDial 2.0. Experiment results show that the use of additional English data can bring performance improvement for Chinese conversational recommendation, indicating the benefits of DuRecDial 2.0. Finally, this dataset provides a challenging testbed for future studies of monolingual, multilingual, and cross-lingual conversational recommendation.
Learning objectives of recommender models remain largely unexplored. Most methods routinely adopt either pointwise or pairwise loss to train the model parameters, while rarely pay attention to softmax loss due to the high computational cost. Sampled softmax loss emerges as an efficient substitute for softmax loss. Its special case, InfoNCE loss, has been widely used in self-supervised learning and exhibited remarkable performance for contrastive learning. Nonetheless, limited studies use sampled softmax loss as the learning objective to train the recommender. Worse still, none of them explore its properties and answer "Does sampled softmax loss suit for item recommendation?" and "What are the conceptual advantages of sampled softmax loss, as compared with the prevalent losses?", to the best of our knowledge. In this work, we aim to better understand sampled softmax loss for item recommendation. Specifically, we first theoretically reveal three model-agnostic advantages: (1) mitigating popularity bias, which is beneficial to long-tail recommendation; (2) mining hard negative samples, which offers informative gradients to optimize model parameters; and (3) maximizing the ranking metric, which facilitates top-K performance. Moreover, we probe the model-specific characteristics on the top of various recommenders. Experimental results suggest that sampled softmax loss is more friendly to history and graph-based recommenders (e.g., SVD++ and LightGCN), but performs poorly for ID-based models (e.g., MF). We ascribe this to its shortcoming in learning representation magnitude, making the combination with the models that are also incapable of adjusting representation magnitude learn poor representations. In contrast, the history- and graph-based models, which naturally adjust representation magnitude according to node degree, are able to compensate for the shortcoming of sampled softmax loss.
Session-based recommendation targets next-item prediction by exploiting user behaviors within a short time period. Compared with other recommendation paradigms, session-based recommendation suffers more from the problem of data sparsity due to the very limited short-term interactions. Self-supervised learning, which can discover ground-truth samples from the raw data, holds vast potentials to tackle this problem. However, existing self-supervised recommendation models mainly rely on item/segment dropout to augment data, which are not fit for session-based recommendation because the dropout leads to sparser data, creating unserviceable self-supervision signals. In this paper, for informative session-based data augmentation, we combine self-supervised learning with co-training, and then develop a framework to enhance session-based recommendation. Technically, we first exploit the session-based graph to augment two views that exhibit the internal and external connectivities of sessions, and then we build two distinct graph encoders over the two views, which recursively leverage the different connectivity information to generate ground-truth samples to supervise each other by contrastive learning. In contrast to the dropout strategy, the proposed self-supervised graph co-training preserves the complete session information and fulfills genuine data augmentation. Extensive experiments on multiple benchmark datasets show that, session-based recommendation can be remarkably enhanced under the regime of self-supervised graph co-training, achieving the state-of-the-art performance.
News recommendation aims to help online news platform users find their preferred news articles. Existing news recommendation methods usually learn models from historical user behaviors on news. However, these behaviors are usually biased on news providers. Models trained on biased user data may capture and even amplify the biases on news providers, and are unfair for some minority news providers. In this paper, we propose a provider fairness-aware news recommendation framework (named ProFairRec), which can learn news recommendation models fair for different news providers from biased user data. The core idea of ProFairRec is to learn provider-fair news representations and provider-fair user representations to achieve provider fairness. To learn provider-fair representations from biased data, we employ provider-biased representations to inherit provider bias from data. Provider-fair and -biased news representations are learned from news content and provider IDs respectively, which are further aggregated to build fair and biased user representations based on user click history. All of these representations are used in model training while only fair representations are used for user-news matching to achieve fair news recommendation. Besides, we propose an adversarial learning task on news provider discrimination to prevent provider-fair news representation from encoding provider bias. We also propose an orthogonal regularization on provider-fair and -biased representations to better reduce provider bias in provider-fair representations. Moreover, ProFairRec is a general framework and can be applied to different news recommendation methods. Extensive experiments on a public dataset verify that our ProFairRec approach can effectively improve the provider fairness of many existing methods and meanwhile maintain their recommendation accuracy.
Visual embeddings from Convolutional Neural Networks (CNN) trained on the ImageNet dataset for the ILSVRC challenge have shown consistently good performance for transfer learning and are widely used in several tasks, including image recommendation. However, some important questions have not yet been answered in order to use these embeddings for a larger scope of recommendation domains: a) Do CNNs that perform better in ImageNet are also better for transfer learning in content-based image recommendation?, b) Does fine-tuning help to improve performance? and c) Which is the best way to perform the fine-tuning? In this paper we compare several CNN models pre-trained with ImageNet to evaluate their transfer learning performance to an artwork image recommendation task. Our results indicate that models with better performance in the ImageNet challenge do not always imply better transfer learning for recommendation tasks (e.g. NASNet vs. ResNet). Our results also show that fine-tuning can be helpful even with a small dataset, but not every fine-tuning works. Our results can inform other researchers and practitioners on how to train their CNNs for better transfer learning towards image recommendation systems.
A sequential recommender system aims to recommend attractive items to users based on behaviour patterns. The predominant sequential recommendation models are based on natural language processing models, such as the gated recurrent unit, that embed items in some defined space and grasp the user's long-term and short-term preferences based on the item embeddings. However, these approaches lack fundamental insight into how such models are related to the user's inherent decision-making process. To provide this insight, we propose a novel recurrent cell, namely FaNC, from Freudian and Newtonian perspectives. FaNC divides the user's state into conscious and unconscious states, and the user's decision process is modelled by Freud's two principles: the pleasure principle and reality principle. To model the pleasure principle, i.e., free-floating user's instinct, we place the user's unconscious state and item embeddings in the same latent space and subject them to Newton's law of gravitation. Moreover, to recommend items to users, we model the reality principle, i.e., balancing the conscious and unconscious states, via a gating function. Based on extensive experiments on various benchmark datasets, this paper provides insight into the characteristics of the proposed model. FaNC initiates a new direction of sequential recommendations at the convergence of psychoanalysis and recommender systems.
Recommender systems (RSs) are becoming an inseparable part of our everyday lives. They help us find our favorite items to purchase, our friends on social networks, and our favorite movies to watch. Traditionally, the recommendation problem was considered as a simple classification or prediction problem; however, the sequential nature of the recommendation problem has been shown. Accordingly, it can be formulated as a Markov decision process (MDP) and reinforcement learning (RL) methods can be employed to solve it. In fact, recent advances in combining deep learning with traditional RL methods, i.e. deep reinforcement learning (DRL), has made it possible to apply RL to the recommendation problem with massive state and action spaces. In this paper, a survey on reinforcement learning based recommender systems (RLRSs) is presented. We first recognize the fact that algorithms developed for RLRSs can be generally classified into RL- and DRL-based methods. Then, we present these RL- and DRL-based methods in a classified manner based on the specific RL algorithm, e.g., Q-learning, SARSA, and REINFORCE, that is used to optimize the recommendation policy. Furthermore, some tables are presented that contain detailed information about the MDP formulation of these methods, as well as about their evaluation schemes. Finally, we discuss important aspects and challenges that can be addressed in the future.
Information retrieval (IR) systems often leverage query data to suggest relevant items to users. This introduces the possibility of unfairness if the query (i.e., input) and the resulting recommendations unintentionally correlate with latent factors that are protected variables (e.g., race, gender, and age). For instance, a visual search system for fashion recommendations may pick up on features of the human models rather than fashion garments when generating recommendations. In this work, we introduce a statistical test for "distribution parity" in the top-K IR results, which assesses whether a given set of recommendations is fair with respect to a specific protected variable. We evaluate our test using both simulated and empirical results. First, using artificially biased recommendations, we demonstrate the trade-off between statistically detectable bias and the size of the search catalog. Second, we apply our test to a visual search system for fashion garments, specifically testing for recommendation bias based on the skin tone of fashion models. Our distribution parity test can help ensure that IR systems' results are fair and produce a good experience for all users.
Multimedia content is of predominance in the modern Web era. Investigating how users interact with multimodal items is a continuing concern within the rapid development of recommender systems. The majority of previous work focuses on modeling user-item interactions with multimodal features included as side information. However, this scheme is not well-designed for multimedia recommendation. Specifically, only collaborative item-item relationships are implicitly modeled through high-order item-user-item relations. Considering that items are associated with rich contents in multiple modalities, we argue that the latent item-item structures underlying these multimodal contents could be beneficial for learning better item representations and further boosting recommendation. To this end, we propose a LATent sTructure mining method for multImodal reCommEndation, which we term LATTICE for brevity. To be specific, in the proposed LATTICE model, we devise a novel modality-aware structure learning layer, which learns item-item structures for each modality and aggregates multiple modalities to obtain latent item graphs. Based on the learned latent graphs, we perform graph convolutions to explicitly inject high-order item affinities into item representations. These enriched item representations can then be plugged into existing collaborative filtering methods to make more accurate recommendations. Extensive experiments on three real-world datasets demonstrate the superiority of our method over state-of-the-art multimedia recommendation methods and validate the efficacy of mining latent item-item relationships from multimodal features.