One of the key challenges of learning an online recommendation model is the temporal domain shift, which causes the mismatch between the training and testing data distribution and hence domain generalization error. To overcome, we propose to learn a meta future gradient generator that forecasts the gradient information of the future data distribution for training so that the recommendation model can be trained as if we were able to look ahead at the future of its deployment. Compared with Batch Update, a widely used paradigm, our theory suggests that the proposed algorithm achieves smaller temporal domain generalization error measured by a gradient variation term in a local regret. We demonstrate the empirical advantage by comparing with various representative baselines.

Sequential recommendation predicts users' next behaviors with their historical interactions. Recommending with longer sequences improves recommendation accuracy and increases the degree of personalization. As sequences get longer, existing works have not yet addressed the following two main challenges. Firstly, modeling long-range intra-sequence dependency is difficult with increasing sequence lengths. Secondly, it requires efficient memory and computational speeds. In this paper, we propose a Sparse Attentive Memory (SAM) network for long sequential user behavior modeling. SAM supports efficient training and real-time inference for user behavior sequences with lengths on the scale of thousands. In SAM, we model the target item as the query and the long sequence as the knowledge database, where the former continuously elicits relevant information from the latter. SAM simultaneously models target-sequence dependencies and long-range intra-sequence dependencies with O(L) complexity and O(1) number of sequential updates, which can only be achieved by the self-attention mechanism with O(L^2) complexity. Extensive empirical results demonstrate that our proposed solution is effective not only in long user behavior modeling but also on short sequences modeling. Implemented on sequences of length 1000, SAM is successfully deployed on one of the largest international E-commerce platforms. This inference time is within 30ms, with a substantial 7.30% click-through rate improvement for the online A/B test. To the best of our knowledge, it is the first end-to-end long user sequence modeling framework that models intra-sequence and target-sequence dependencies with the aforementioned degree of efficiency and successfully deployed on a large-scale real-time industrial recommender system.

In this work we introduce an incremental learning framework for Click-Through-Rate (CTR) prediction and demonstrate its effectiveness for Taboola's massive-scale recommendation service. Our approach enables rapid capture of emerging trends through warm-starting from previously deployed models and fine tuning on "fresh" data only. Past knowledge is maintained via a teacher-student paradigm, where the teacher acts as a distillation technique, mitigating the catastrophic forgetting phenomenon. Our incremental learning framework enables significantly faster training and deployment cycles (x12 speedup). We demonstrate a consistent Revenue Per Mille (RPM) lift over multiple traffic segments and a significant CTR increase on newly introduced items.

Most of your day is probably spent on your phone. From sending emails to watching videos, from searching for work to reading the news, it's almost impossible not to spend time on one of the most important pieces of technology in our world. That's why it was only a matter of time until artificial intelligence came knocking on that door. Though sometimes, as with any new technology, all the possibilities and potential greatness can be somewhat overwhelming. Think of artificial intelligence as technology that can learn, understand, and make decisions and predictions similar to a human.

Contextual multi-armed bandit (MAB) is an important sequential decision-making problem in recommendation systems. A line of works, called the clustering of bandits (CLUB), utilize the collaborative effect over users and dramatically improve the recommendation quality. Owing to the increasing application scale and public concerns about privacy, there is a growing demand to keep user data decentralized and push bandit learning to the local server side. Existing CLUB algorithms, however, are designed under the centralized setting where data are available at a central server. We focus on studying the federated online clustering of bandit (FCLUB) problem, which aims to minimize the total regret while satisfying privacy and communication considerations. We design a new phase-based scheme for cluster detection and a novel asynchronous communication protocol for cooperative bandit learning for this problem. To protect users' privacy, previous differential privacy (DP) definitions are not very suitable, and we propose a new DP notion that acts on the user cluster level. We provide rigorous proofs to show that our algorithm simultaneously achieves (clustered) DP, sublinear communication complexity and sublinear regret. Finally, experimental evaluations show our superior performance compared with benchmark algorithms.

The mission of Just Eat Takeaway is to empower users' every food moment. A big part of fulfilling that mission is ensuring that we show the right restaurants to the right users. To do this, we design large-scale machine learning recommendation systems that can learn which types of users tend to enjoy which types of restaurants. A crucial part of this process is being able to compare the quality of different recommendation systems, in order to decide which is most effective. The gold standard approach to this evaluation problem is A/B testing - allow the different systems to recommend items to randomly selected groups of users, and compare their relative performance. However, A/B testing is both slow, taking time to reach sufficient statistical power, and expensive [5, 1] if we serve poor recommendations to a subset of users. Therefore, evaluating recommender systems in offline fashion without running A/B tests is of significant importance. These offline approaches avoid serving experimental recommenders to users, and instead evaluates them using old online data, where users were served recommendations from a different system. Metrics such as Root Mean Squared Error [2] Mean Average Precision, Normalized Discounted Cumulative Gain, Mean Reciprocal Rank and Hit Rate have been used repeatedly for offline evaluation.

SHOPPING: Products featured in this article are independently selected by our shopping writers. If you make a purchase using links on this page, MailOnline will earn an affiliate commission. Amazon has launched a massive End of Summer Sale with unmissable discounts on their bestselling Echo devices. But one deal not to be missed is the Amazon Echo 4th Gen Smart Bluetooth Speaker, which usually retails for £49.99, is currently on sale for £22.99. It's a must-have if you've been considering buying the popular gadget for some time or want to automate your home.

For years, billions in venture capital has poured into fintech banks like Chime and N26 on the bet such upstarts can wrest away the lion's share of an estimated $469 trillion in assets held globally by other financial institutions and retail banks. Banks have held their own through the pandemic, reporting record 2021 profits on low chargeoff rates, rising customer deposits and thriving investment opportunities. Yet a new survey of 142 banking executives around the world, conducted by Capgemini and Qorus for the World Retail Banking Report 2022, found that 70% of them believe they lack foundational data analysis and AI capabilities to compete long term. The technology empowering decentralised finance – where consumers bank when and where they want – is now augmented with a more sophisticated, AI-driven banking experience. Mobile apps enable more than just bill pay as AI-infused virtual assistants alert customers to potential fraudulent activity or transfer money via voice commands.

In one-class recommendation systems, the goal is to learn a model from a small set of interacted users and items and then identify the positively-related user-item pairs among a large number of pairs with unknown interactions. Most previous loss functions rely on dissimilar pairs of users and items, which are selected from the ones with unknown interactions, to obtain better prediction performance. This strategy introduces several challenges such as increasing training time and hurting the performance by picking "similar pairs with the unknown interactions" as dissimilar pairs. In this paper, the goal is to only use the similar set to train the models. We point out three trivial solutions that the models converge to when they are trained only on similar pairs: collapsed, partially collapsed, and shrinking solutions. We propose two terms that can be added to the objective functions in the literature to avoid these solutions. The first one is a hinge pairwise distance loss that avoids the shrinking and collapsed solutions by keeping the average pairwise distance of all the representations greater than a margin. The second one is an orthogonality term that minimizes the correlation between the dimensions of the representations and avoids the partially collapsed solution. We conduct experiments on a variety of tasks on public and real-world datasets. The results show that our approach using only similar pairs outperforms state-of-the-art methods using similar pairs and a large number of dissimilar pairs.

While machine learning and ranking-based systems are in widespread use for sensitive decision-making processes (e.g., determining job candidates, assigning credit scores), they are rife with concerns over unintended biases in their outcomes, which makes algorithmic fairness (e.g., demographic parity, equal opportunity) an objective of interest. 'Algorithmic recourse' offers feasible recovery actions to change unwanted outcomes through the modification of attributes. We introduce the notion of ranked group-level recourse fairness, and develop a 'recourse-aware ranking' solution that satisfies ranked recourse fairness constraints while minimizing the cost of suggested modifications. Our solution suggests interventions that can reorder the ranked list of database records and mitigate group-level unfairness; specifically, disproportionate representation of sub-groups and recourse cost imbalance. This re-ranking identifies the minimum modifications to data points, with these attribute modifications weighted according to their ease of recourse. We then present an efficient block-based extension that enables re-ranking at any granularity (e.g., multiple brackets of bank loan interest rates, multiple pages of search engine results). Evaluation on real datasets shows that, while existing methods may even exacerbate recourse unfairness, our solution -- RAGUEL -- significantly improves recourse-aware fairness. RAGUEL outperforms alternatives at improving recourse fairness, through a combined process of counterfactual generation and re-ranking, whilst remaining efficient for large-scale datasets.