Recent methods utilize graph contrastive Learning within graph-structured user-item interaction data for collaborative filtering and have demonstrated their efficacy in recommendation tasks. However, they ignore that the difference relation density of nodes between the user- and item-side causes the adaptability of graphs on bilateral nodes to be different after multi-hop graph interaction calculation, which limits existing models to achieve ideal results. To solve this issue, we propose a novel framework for recommendation tasks called Bilateral Unsymmetrical Graph Contrastive Learning (BusGCL) that consider the bilateral unsymmetry on user-item node relation density for sliced user and item graph reasoning better with bilateral slicing contrastive training. Especially, taking into account the aggregation ability of hypergraph-based graph convolutional network (GCN) in digging implicit similarities is more suitable for user nodes, embeddings generated from three different modules: hypergraph-based GCN, GCN and perturbed GCN, are sliced into two subviews by the user- and item-side respectively, and selectively combined into subview pairs bilaterally based on the characteristics of inter-node relation structure. Furthermore, to align the distribution of user and item embeddings after aggregation, a dispersing loss is leveraged to adjust the mutual distance between all embeddings for maintaining learning ability. Comprehensive experiments on two public datasets have proved the superiority of BusGCL in comparison to various recommendation methods. Other models can simply utilize our bilateral slicing contrastive learning to enhance recommending performance without incurring extra expenses.

Research has indicated that it is often more convenient, faster, and cost-effective to gather feedback in a relative manner rather than using absolute ratings [31, 40]. To illustrate, when assessing an individual's preference between two items, such as A and B, it is often easier for respondents to answer preference-oriented queries like "Which item do you prefer, A or B?" instead of requesting to rate items A and B on a scale ranging from 0 to 10. From the perspective of a system designer, leveraging this user preference data can significantly enhance system performance, especially when this data can be collected in a relative and online fashion. This applies to various real-world scenarios, including recommendation systems, crowd-sourcing platforms, training bots, multiplayer games, search engine optimization, online retail, and more. In many practical situations, particularly when human preferences are gathered online, such as designing surveys, expert reviews, product selection, search engine optimization, recommender systems, multiplayer game rankings, and even broader reinforcement learning problems with complex reward structures, it's often easier to elicit preference feedback instead of relying on absolute ratings or rewards. Because of its broad utility and the simplicity of gathering data using relative feedback, learning from preferences has become highly popular in the machine learning community. It has been extensively studied over the past decade under the name "Dueling-Bandits" (DB) in the literature. This framework is an extension of the traditional multi-armed bandit (MAB) setting, as described in [4]. In the DB framework, the goal is to identify a set of'good' options from a fixed decision

A New Jersey couple woke up to a 67-minute-long voicemail from an'unknown caller' - and discovered it was left by their Amazon Alexa. 'I was checking the message ... and was like, wait, this is me talking in the bedroom,' she said. Alexa can call your smartphone if you trigger the'Find My Phone' feature, but a company spokesperson said the Amazon Echo doesn't record or store conversations unless it hears the'wake word,' prompting a light on the device to turn on to let you know it's listening. Amazon has come under fire for its devices recording conversations and faced two separate privacy violation lawsuits last year, including a claim that it had violated children's privacy rights by refusing to remove the recording history of minors. A judge ruled that the company had to pay out a collective 30.8 million for both violations. 'There wasn't a lot of talking in the message, mostly bleeping,' Creegan said, but added that she could hear snippets of her telling Alexa to'turn the lights off' adding that there was'two or three sentences of me talking to the dog.

Training recommendation models pose significant challenges regarding resource utilization and performance. Prior research has proposed an approach that categorizes embeddings into popular and non-popular classes to reduce the training time for recommendation models. We observe that, even among the popular embeddings, certain embeddings undergo rapid training and exhibit minimal subsequent variation, resulting in saturation. Consequently, updates to these embeddings lack any contribution to model quality. This paper presents Slipstream, a software framework that identifies stale embeddings on the fly and skips their updates to enhance performance. This capability enables Slipstream to achieve substantial speedup, optimize CPU-GPU bandwidth usage, and eliminate unnecessary memory access. SlipStream showcases training time reductions of 2x, 2.4x, 1.2x, and 1.175x across real-world datasets and configurations, compared to Baseline XDL, Intel-optimized DRLM, FAE, and Hotline, respectively.

Recommendation systems, as widely implemented nowadays on various platforms, recommend relevant items to users based on their preferences. The classical methods which rely on user-item interaction matrices has limitations, especially in scenarios where there is a lack of interaction data for new items. Knowledge graph (KG)-based recommendation systems have emerged as a promising solution. However, most KG-based methods adopt node embeddings, which do not provide personalized recommendations for different users and cannot generalize well to the new items. To address these limitations, we propose Knowledge-enhanced User-Centric subgraph Network (KUCNet), a subgraph learning approach with graph neural network (GNN) for effective recommendation. KUCNet constructs a U-I subgraph for each user-item pair that captures both the historical information of user-item interactions and the side information provided in KG. An attention-based GNN is designed to encode the U-I subgraphs for recommendation. Considering efficiency, the pruned user-centric computation graph is further introduced such that multiple U-I subgraphs can be simultaneously computed and that the size can be pruned by Personalized PageRank. Our proposed method achieves accurate, efficient, and interpretable recommendations especially for new items. Experimental results demonstrate the superiority of KUCNet over state-of-the-art KG-based and collaborative filtering (CF)-based methods.

This paper investigates statistical inference for noisy matrix completion in a semi-supervised model when auxiliary covariates are available. The model consists of two parts. One part is a low-rank matrix induced by unobserved latent factors; the other part models the effects of the observed covariates through a coefficient matrix which is composed of high-dimensional column vectors. We model the observational pattern of the responses through a logistic regression of the covariates, and allow its probability to go to zero as the sample size increases. We apply an iterative least squares (LS) estimation approach in our considered context. The iterative LS methods in general enjoy a low computational cost, but deriving the statistical properties of the resulting estimators is a challenging task. We show that our method only needs a few iterations, and the resulting entry-wise estimators of the low-rank matrix and the coefficient matrix are guaranteed to have asymptotic normal distributions. As a result, individual inference can be conducted for each entry of the unknown matrices. We also propose a simultaneous testing procedure with multiplier bootstrap for the high-dimensional coefficient matrix. This simultaneous inferential tool can help us further investigate the effects of covariates for the prediction of missing entries.

The Amazon Spring Sale is here and if you're interested in tech deals, you've come to the right place. However, don't mistake this for a spring Prime Day -- unlike Amazon's bigger, traditional sale events, this one doesn't revolve around Prime-exclusive discounts. And that's a good thing; that means anyone who shops on Amazon can take advantage of the deals. Given the seasonal nature of this event, it's not a boon for discounts on laptops, tablets, wearables and the like. However, we were able to find a number of decent discounts worth your time and money. While most of these Amazon deal prices are not the same as those we saw around Black Friday last year, some get pretty close (as a general rule of thumb, a good price in March isn't necessarily the same thing as a good price in November). Here are the best Amazon Big Spring Sale deals on tech we love that you can get right now. Apple's AirPods Pro are once again available for 189, which matches the best price we've seen for the latest iteration with a USB-C charging case. Apple normally sells the noise-canceling earphones for 249, though we often see them go closer to 200 at third-party retailers. Either way, they remain our favorite wireless earbuds for iOS users, as they provide an array of perks when paired with an iPhone, from faster pairing to hands-free Siri. Their battery life and mic quality are just OK these days, but this pair should serve you well if you're all-in on Apple.

The Amazon Big Spring Sale is in full swing, and one of our favorite affordable pairs of wireless earbuds is even cheaper because of it. The 2023 Echo Buds are down to 35 in this Amazon deal, which is their lowest price yet. These Echo Buds have a lot of improvements over the previous model, and we like them for their detailed and balanced sound profile, built-in Alexa support and five hours of battery life. It's also worth noting that this deal is available to anyone. Amazon's Echo Buds are available today for a record-low price.

This smart lightbulb from Kasa requires no hub or special equipment to operate. Set timers, on/off schedules, and fully adjust the brightness from 1-100% using an app, or with your voice though Amazon Alexa or Google Assistant. The list price on this bulb is 16.99, but right now on Amazon you can buy it for just 7.99, the lowest price in over a year. That's a great deal if you're looking to add some basic automation to the lights inside your home, or outside on a porch. The 53% discount is active on Amazon right now (see it here), where the bulb gets 4.5 out of 5 stars from nearly 15,000 reviewers.

User embeddings play a crucial role in user engagement forecasting and personalized services. Recent advances in sequence modeling have sparked interest in learning user embeddings from behavioral data. Yet behavior-based user embedding learning faces the unique challenge of dynamic user modeling. As users continuously interact with the apps, user embeddings should be periodically updated to account for users' recent and long-term behavior patterns. Existing methods highly rely on stateless sequence models that lack memory of historical behavior. They have to either discard historical data and use only the most recent data or reprocess the old and new data jointly. Both cases incur substantial computational overhead. To address this limitation, we introduce User Stateful Embedding (USE). USE generates user embeddings and reflects users' evolving behaviors without the need for exhaustive reprocessing by storing previous model states and revisiting them in the future. Furthermore, we introduce a novel training objective named future W-behavior prediction to transcend the limitations of next-token prediction by forecasting a broader horizon of upcoming user behaviors. By combining it with the Same User Prediction, a contrastive learning-based objective that predicts whether different segments of behavior sequences belong to the same user, we further improve the embeddings' distinctiveness and representativeness. We conducted experiments on 8 downstream tasks using Snapchat users' behavioral logs in both static (i.e., fixed user behavior sequences) and dynamic (i.e., periodically updated user behavior sequences) settings. We demonstrate USE's superior performance over established baselines. The results underscore USE's effectiveness and efficiency in integrating historical and recent user behavior sequences into user embeddings in dynamic user modeling.