Graph neural networks (GNNs) have recently emerged as an effective approach to model neighborhood signals in collaborative filtering. Towards this research line, graph contrastive learning (GCL) demonstrates robust capabilities to address the supervision label shortage issue through generating massive self-supervised signals. Despite its effectiveness, GCL for recommendation suffers seriously from two main challenges: i) GCL relies on graph augmentation to generate semantically different views for contrasting, which could potentially disrupt key information and introduce unwanted noise; ii) current works for GCL primarily focus on contrasting representations using sophisticated networks architecture (usually deep) to capture high-order interactions, which leads to increased computational complexity and suboptimal training efficiency. To this end, we propose L2CL, a principled Layer-to-Layer Contrastive Learning framework that contrasts representations from different layers. By aligning the semantic similarities between different layers, L2CL enables the learning of complex structural relationships and gets rid of the noise perturbation in stochastic data augmentation. Surprisingly, we find that L2CL, using only one-hop contrastive learning paradigm, is able to capture intrinsic semantic structures and improve the quality of node representation, leading to a simple yet effective architecture. We also provide theoretical guarantees for L2CL in minimizing task-irrelevant information. Extensive experiments on five real-world datasets demonstrate the superiority of our model over various state-of-the-art collaborative filtering methods. Our code is available at https://github.com/downeykking/L2CL.

If you ever feel like Gen Z is speaking an entirely different language, you are definitely not alone. Luckily for all the baffled parents out there, a new Alexa update should help you avoid giving your kids the ick with your sus Gen Z slang. From today, Amazon users will be able to ask Alexa to'talk Gen Z to me' to reveal one of 20 phrases using Gen Z's favourite words. That might include phrases such as'the math isn't mathing' which Alexa defines as'something is incorrect or unreasonable, something doesn't add up or make sense'. And as Amazon's survey of 2,385 Britons reveals that 83 per cent of Gen Z say the older generation fails to understand them, a little extra assistance might be welcome.

Prime Day is officially over. Did your friend mention a killer deal they scored? Are you now dealing with FOMO? Well not to worry, roughly half of the Amazon Prime Day deals we highlighted during the main event are still kicking around, though they are expiring quickly. These are all products we here at WIRED have tested and recommend--some prices have slightly increased but are still a sale price, while a few have gone lower. Your next opportunity to score a good deal is around October and November, for Amazon's second Prime Day sale event and Black Friday, so take advantage, but only buy something if you actually want or need it. We test products year-round and handpicked these Prime Day deals. Products that are sold out or no longer discounted will be crossed out. We'll update this guide regularly throughout Prime Day by adding fresh deals and removing dead deals. If you buy something using links in our stories, we may earn a commission. This helps support our journalism.

Amazon Prime Day is officially over -- but there are still a few legacy deals hanging on. We had previously pulled together this list of worthy under- 25 deals, and we've just updated it to reflect these affordable bargains that are still available as of Thursday morning. As a reminder (and for the uninitiated): Engadget treats tech deals with the same care as we would "regular" tech news. When we scour the web for deals, we're looking not only for the best prices possible, but also the best products as well. Our goal with our deals coverage, especially surrounding events like Amazon Prime Day, is to surface only the best deals we can find on the gadgets we've tested and rated highly, or that we've used and know to be worth your money.

Despite its breakthrough in classification problems, Knowledge distillation (KD) to recommendation models and ranking problems has not been studied well in the previous literature. This dissertation is devoted to developing knowledge distillation methods for recommender systems to fully improve the performance of a compact model. We propose novel distillation methods designed for recommender systems. The proposed methods are categorized according to their knowledge sources as follows: (1) Latent knowledge: we propose two methods that transfer latent knowledge of user/item representation. They effectively transfer knowledge of niche tastes with a balanced distillation strategy that prevents the KD process from being biased towards a small number of large preference groups. Also, we propose a new method that transfers user/item relations in the representation space. The proposed method selectively transfers essential relations considering the limited capacity of the compact model. (2) Ranking knowledge: we propose three methods that transfer ranking knowledge from the recommendation results. They formulate the KD process as a ranking matching problem and transfer the knowledge via a listwise learning strategy. Further, we present a new learning framework that compresses the ranking knowledge of heterogeneous recommendation models. The proposed framework is developed to ease the computational burdens of model ensemble which is a dominant solution for many recommendation applications. We validate the benefit of our proposed methods and frameworks through extensive experiments. To summarize, this dissertation sheds light on knowledge distillation approaches for a better accuracy-efficiency trade-off of the recommendation models.

Offline reinforcement learning (RL) is an effective tool for real-world recommender systems with its capacity to model the dynamic interest of users and its interactive nature. Most existing offline RL recommender systems focus on model-based RL through learning a world model from offline data and building the recommendation policy by interacting with this model. Although these methods have made progress in the recommendation performance, the effectiveness of model-based offline RL methods is often constrained by the accuracy of the estimation of the reward model and the model uncertainties, primarily due to the extreme discrepancy between offline logged data and real-world data in user interactions with online platforms. To fill this gap, a more accurate reward model and uncertainty estimation are needed for the model-based RL methods. In this paper, a novel model-based Reward Shaping in Offline Reinforcement Learning for Recommender Systems, ROLeR, is proposed for reward and uncertainty estimation in recommendation systems. Specifically, a non-parametric reward shaping method is designed to refine the reward model. In addition, a flexible and more representative uncertainty penalty is designed to fit the needs of recommendation systems. Extensive experiments conducted on four benchmark datasets showcase that ROLeR achieves state-of-the-art performance compared with existing baselines. The source code can be downloaded at https://github.com/ArronDZhang/ROLeR.

Recommender Systems (RS) play an integral role in enhancing user experiences by providing personalized item suggestions. This survey reviews the progress in RS inclusively from 2017 to 2024, effectively connecting theoretical advances with practical applications. We explore the development from traditional RS techniques like content-based and collaborative filtering to advanced methods involving deep learning, graph-based models, reinforcement learning, and large language models. We also discuss specialized systems such as context-aware, review-based, and fairness-aware RS. The primary goal of this survey is to bridge theory with practice. It addresses challenges across various sectors, including e-commerce, healthcare, and finance, emphasizing the need for scalable, real-time, and trustworthy solutions. Through this survey, we promote stronger partnerships between academic research and industry practices. The insights offered by this survey aim to guide industry professionals in optimizing RS deployment and to inspire future research directions, especially in addressing emerging technological and societal trends

This study examines the performance of item-based k-Nearest Neighbors (ItemKNN) algorithms in the RecBole and LensKit recommender system libraries. Using four data sets (Anime, Modcloth, ML-100K, and ML-1M), we assess each library's efficiency, accuracy, and scalability, focusing primarily on normalized discounted cumulative gain (nDCG). Our results show that RecBole outperforms LensKit on two of three metrics on the ML-100K data set: it achieved an 18% higher nDCG, 14% higher precision, and 35% lower recall. To ensure a fair comparison, we adjusted LensKit's nDCG calculation to match RecBole's method. This alignment made the performance more comparable, with LensKit achieving an nDCG of 0.2540 and RecBole 0.2674. Differences in similarity matrix calculations were identified as the main cause of performance deviations. After modifying LensKit to retain only the top K similar items, both libraries showed nearly identical nDCG values across all data sets. For instance, both achieved an nDCG of 0.2586 on the ML-1M data set with the same random seed. Initially, LensKit's original implementation only surpassed RecBole in the ModCloth dataset.

On the second day of Amazon's Prime Day sale, the deals on smaller gadgets and accessories are still going strong. In fact, as I was checking to make sure these deals were still live, I noted about five on the list that dropped a few dollars cheaper than they were yesterday. As a reminder, this list represents the best of the affordable tech gear that we at Engadget have tested, reviewed and know to be worth your time. Everything here is on sale for 49.99 or under to make up the best possible roundup of the Prime Day tech deals you can get for under 50. As with all Engadget tech deals coverage, we only highlight discounts on gear we've tested or have otherwise used and know to be worthy of your money. We cross-checked our guides and reviews with the Prime Day deals Amazon has put forth to come up with what you see here. The Anker Nano power bank in black is on sale for Prime Day for 16.13. That's a 15 percent discount and a good deal for one of the best power banks we tested. We like the foldable USB-C connector which means you don't have to remember a separate cable and the amount of charge it delivers for such a small package. Plus it's compact enough you can use it while it refills your phone.

Apple's new iOS update is just weeks away - and more details are starting to emerge about what is shaping up to be the biggest update yet. Originally announced at the company's annual Worldwide Developers Conference (WWDC) in June, the software is expected to hit devices in mid-September. Apple is touting its AI integration with iOS 18, iPadOS 18, and macOS Sequoia - called'Apple Intelligence' - as the update's flagship feature. The feature will bring a lot of change to the operating system, including new-and-improved Siri capabilities, the ability to create custom emojis, and more. But the change that made headlines most was a new app-hiding feature dubbed'cheater's paradise' because it allows people to conceal certain apps (like dating apps, for instance) in locked folders.