In many scenarios, recommender system user interaction data such as clicks or ratings is sparse, and item turnover rates (e.g., new articles, job postings) high. Given this, the integration of contextual "side" information in addition to user-item ratings is highly desirable. Whilst there are algorithms that can handle both rating and contextual data simultaneously, these algorithms are typically limited to making only in-sample recommendations, suffer from the curse of dimensionality, and do not incorporate multi-armed bandit (MAB) policies for long-term cumulative reward optimization. We propose multi-view interactive topic regression (MV-ICTR) a novel partially online latent factor recommender algorithm that incorporates both rating and contextual information to model item-specific feature dependencies and users' personal preferences simultaneously, with multi-armed bandit policies for continued online personalization. The result is significantly increased performance on datasets with high percentages of cold-start users and items.

The cold start problem is a common challenge in various domains, including media use cases such as predicting viewership for newly launched shows on Over-The-Top (OTT) platforms. In this study, we propose a generic approach to tackle cold start problems by leveraging metadata and employing multi-model ensemble techniques. Our methodology includes feature engineering, model selection, and an ensemble approach based on a weighted average of predictions. The performance of our proposed method is evaluated using various performance metrics. Our results indicate that the multi-model ensemble approach significantly improves prediction accuracy compared to individual models.

Deep learning-based recommendation systems (e.g., DLRMs) are widely used AI models to provide high-quality personalized recommendations. Training data used for modern recommendation systems commonly includes categorical features taking on tens-of-millions of possible distinct values. These categorical tokens are typically assigned learned vector representations, that are stored in large embedding tables, on the order of 100s of GB. Storing and accessing these tables represent a substantial burden in commercial deployments. Our work proposes MEM-REC, a novel alternative representation approach for embedding tables. MEM-REC leverages bloom filters and hashing methods to encode categorical features using two cache-friendly embedding tables. The first table (token embedding) contains raw embeddings (i.e. learned vector representation), and the second table (weight embedding), which is much smaller, contains weights to scale these raw embeddings to provide better discriminative capability to each data point. We provide a detailed architecture, design and analysis of MEM-REC addressing trade-offs in accuracy and computation requirements, in comparison with state-of-the-art techniques. We show that MEM-REC can not only maintain the recommendation quality and significantly reduce the memory footprint for commercial scale recommendation models but can also improve the embedding latency. In particular, based on our results, MEM-REC compresses the MLPerf CriteoTB benchmark DLRM model size by 2900x and performs up to 3.4x faster embeddings while achieving the same AUC as that of the full uncompressed model.

Eugenia Kuyda defended AI companion bots during an interview with Fox News Digital and argued that dating app Replika is just one of many possible solutions to loneliness. AI, or artificial intelligence, is a branch of computer science that is designed to understand and store human intelligence, mimic human capabilities including the completion of tasks, process human language and perform speech recognition. AI is the leading innovation in technology today and its primary goal is to eliminate tedious tasks and assist in immediately accessing extremely detailed and hyper-focused information and data. AI has the ability to consume and process massive datasets and develop patterns to make predictions for the completion of future tasks. While the interest in AI around the world is growing, the science poses an existential crisis for jobs, companies, whole industries and potentially human existence.

Federated recommendation is a new Internet service architecture that aims to provide privacy-preserving recommendation services in federated settings. Existing solutions are used to combine distributed recommendation algorithms and privacy-preserving mechanisms. Thus it inherently takes the form of heavyweight models at the server and hinders the deployment of on-device intelligent models to end-users. This paper proposes a novel Personalized Federated Recommendation (PFedRec) framework to learn many user-specific lightweight models to be deployed on smart devices rather than a heavyweight model on a server. Moreover, we propose a new dual personalization mechanism to effectively learn fine-grained personalization on both users and items. The overall learning process is formulated into a unified federated optimization framework. Specifically, unlike previous methods that share exactly the same item embeddings across users in a federated system, dual personalization allows mild finetuning of item embeddings for each user to generate user-specific views for item representations which can be integrated into existing federated recommendation methods to gain improvements immediately. Experiments on multiple benchmark datasets have demonstrated the effectiveness of PFedRec and the dual personalization mechanism. Moreover, we provide visualizations and in-depth analysis of the personalization techniques in item embedding, which shed novel insights on the design of recommender systems in federated settings. The code is available.

Recent studies have proposed unified user modeling frameworks that leverage user behavior data from various applications. Many of them benefit from utilizing users' behavior sequences as plain texts, representing rich information in any domain or system without losing generality. Hence, a question arises: Can language modeling for user history corpus help improve recommender systems? While its versatile usability has been widely investigated in many domains, its applications to recommender systems still remain underexplored. We show that language modeling applied directly to task-specific user histories achieves excellent results on diverse recommendation tasks. Also, leveraging additional task-agnostic user histories delivers significant performance benefits. We further demonstrate that our approach can provide promising transfer learning capabilities for a broad spectrum of real-world recommender systems, even on unseen domains and services.

The biggest news in tech this week came from Google's annual developer conference on Wednesday. They announced three new devices: The Pixel 7a smartphone, the Pixel Tablet and the Pixel Fold. Discounts on brand new products don't happen often, but both Amazon and Google were quick to bundle Pixel 7a orders with a $50 Amazon gift card, or a free pair of Pixel Buds, respectively -- not sales per se, but free stuff is still compelling for anyone already planning on getting a new phone. Of course, there were deals unrelated to Google too, like savings on Amazon devices including the Fire TV Stick 4K Max, Echo speakers, and nearly all Kindle models. Here are the best tech deals from this week that you can still get today.

Existing recommender systems face difficulties with zero-shot items, i.e. items that have no historical interactions with users during the training stage. Though recent works extract universal item representation via pre-trained language models (PLMs), they ignore the crucial item relationships. This paper presents a novel paradigm for the Zero-Shot Item-based Recommendation (ZSIR) task, which pre-trains a model on product knowledge graph (PKG) to refine the item features from PLMs. We identify three challenges for pre-training PKG, which are multi-type relations in PKG, semantic divergence between item generic information and relations and domain discrepancy from PKG to downstream ZSIR task. We address the challenges by proposing four pre-training tasks and novel task-oriented adaptation (ToA) layers. Moreover, this paper discusses how to fine-tune the model on new recommendation task such that the ToA layers are adapted to ZSIR task. Comprehensive experiments on 18 markets dataset are conducted to verify the effectiveness of the proposed model in both knowledge prediction and ZSIR task.

We develop a novel latent-bandit algorithm for tackling the cold-start problem for new users joining a recommender system. This new algorithm significantly outperforms the state of the art, simultaneously achieving both higher accuracy and lower regret.

Twitter users are calling on Elon Musk to develop an AI-powered dating app called'Twinder,' touting it as the way'to save humanity from extinction.' The idea came after Musk replied'population collapse' to a tweet showing how fertility rates keep dropping in the Nordic countries. The potential dating app, which the Twitter CEO deemed an'interesting idea,' would use artificial intelligence to make matches instead of random swiping. The suggested service would feed AI Twitter accounts, including posts, comments and likes, and the technology would look for another user with similar behaviors and interests. The Twitter thread, viewed over two million times, has hundreds of comments, with some sharing how they met their partner on the social network.