Systematic under-counting effects are observed in data collected across many disciplines, e.g., epidemiology and ecology. Under-counted tensor completion (UC-TC) is well-motivated for many data analytics tasks, e.g., inferring the case numbers of infectious diseases at unobserved locations from under-counted case numbers in neighboring regions. However, existing methods for similar problems often lack supports in theory, making it hard to understand the underlying principles and conditions beyond empirical successes. In this work, a low-rank Poisson tensor model with an expressive unknown nonlinear side information extractor is proposed for under-counted multi-aspect data. A joint low-rank tensor completion and neural network learning algorithm is designed to recover the model. Moreover, the UC-TC formulation is supported by theoretical analysis showing that the fully counted entries of the tensor and each entry's under-counting probability can be provably recovered from partial observations -- under reasonable conditions. To our best knowledge, the result is the first to offer theoretical supports for under-counted multi-aspect data completion. Simulations and real-data experiments corroborate the theoretical claims.

Item-to-Item (I2I) recommendation is an important function in most recommendation systems, which generates replacement or complement suggestions for a particular item based on its semantic similarities to other cataloged items. Given that subsets of items in a recommendation system might be co-interacted with by the same set of customers, graph-based models, such as graph neural networks (GNNs), provide a natural framework to combine, ingest and extract valuable insights from such high-order relational interactions between cataloged items, as well as their metadata features, as has been shown in many recent studies. However, learning GNNs effectively for I2I requires ingesting a large amount of relational data, which might not always be available, especially in new, emerging market segments. To mitigate this data bottleneck, we postulate that recommendation patterns learned from existing mature market segments (with private data) could be adapted to build effective warm-start models for emerging ones. To achieve this, we propose and investigate a personalized federated modeling framework based on GNNs to summarize, assemble and adapt recommendation patterns across market segments with heterogeneous customer behaviors into effective local models. Our key contribution is a personalized graph adaptation model that bridges the gap between recent literature on federated GNNs and (non-graph) personalized federated learning, which either does not optimize for the adaptability of the federated model or is restricted to local models with homogeneous parameterization, excluding GNNs with heterogeneous local graphs.

On E-commerce stores, there are rich recommendation content to help shoppers shopping more efficiently. However given numerous products, it's crucial to select most relevant content to reduce the burden of information overload. We introduced a content ranking service powered by a linear causal bandit algorithm to rank and select content for each shopper under each context. The algorithm mainly leverages aggregated customer behavior features, and ignores single shopper level past activities. We study the problem of inferring shoppers interest from historical activities. We propose a deep learning based bandit algorithm that incorporates historical shopping behavior, customer latent shopping goals, and the correlation between customers and content categories. This model produces more personalized content ranking measured by 12.08% nDCG lift.

Fox News Flash top sports headlines are here. Check out what's clicking on Foxnews.com. At long last, an iPhone finally went off while someone was broadcasting a Tampa Bay Rays game. Because the Rays have a guy named Jose Siri on their team. And yes, his last name is pronounced just like the iPhone's "Siri."

Apparently, the introduction of Windows Copilot signaled the end of Cortana on Microsoft's desktop OS. In a new support document first spotted by Windows Central, the tech giant has announced that it will stop supporting the standalone Cortana app for Windows in late 2023. Microsoft launched Cortana as a voice assistant for Windows mobile devices back in 2014. It was supposed to be the company's answer to Apple's Siri, and it even predates Amazon's Alexa, but it never quite achieved their level of recognition and popularity. Over the year, Microsoft scaled back its plans for the voice assistant until it discontinued its Android and iOS apps back in 2021.

Microsoft launched Cortana as an AI assistant and the flagship feature of Windows 10 in 2015. Now, eight years later, Microsoft is pulling the plug. In a support document, Microsoft said that it's ending support for the Cortana app, Cortana's only remaining presence within Windows. Instead, Microsoft said it will encourage users to use other AI-powered features, whether it be within a standalone app or simply part of Windows or Microsoft Edge. Microsoft did say that Cortana will still be available within Outlook Mobile and various versions of Teams, including Microsoft's conferencing solution, Teams Rooms.

Dear Prudence is Slate's advice column. For this edition, Alicia Montgomery, Slate's vice president of audio, will be filling in as Prudie. My friend Kari and I have been close since we were college roommates (we are now just about 40). Kari has been with her long-distance girlfriend Lora for the last four years, and recently Lora has been talking about moving to Kari and my town in order to better facilitate having a baby. The road for them is going to be long, given the mechanics and their ages, but they have all systems go from their doctors. The problem is that I know Kari is not 100 percent committed to Lora; she says she's not sure she's the one and has built (but not, to my knowledge, deployed) dating profiles on multiple sites and expresses jealousy to me quite often about my adventurous dating life.

Once you have an Alexa-enabled device like an Amazon Echo, open the Alexa app on your smartphone or tablet and go to Settings Alexa Privacy Manage your Alexa Data Choose how long to save recordings. Select "Don't save recordings" and hit confirm. Delete past recordings in the Alexa Privacy section, including your voice history and history of detected sounds.

Recommending novel content, which expands user horizons by introducing them to new interests, has been shown to improve users' long-term experience on recommendation platforms \cite{chen2021values}. Users however are not constantly looking to explore novel content. It is therefore crucial to understand their novelty-seeking intent and adjust the recommendation policy accordingly. Most existing literature models a user's propensity to choose novel content or to prefer a more diverse set of recommendations at individual interactions. Hierarchical structure, on the other hand, exists in a user's novelty-seeking intent, which is manifested as a static and intrinsic user preference for seeking novelty along with a dynamic session-based propensity. To this end, we propose a novel hierarchical reinforcement learning-based method to model the hierarchical user novelty-seeking intent, and to adapt the recommendation policy accordingly based on the extracted user novelty-seeking propensity. We further incorporate diversity and novelty-related measurement in the reward function of the hierarchical RL (HRL) agent to encourage user exploration \cite{chen2021values}. We demonstrate the benefits of explicitly modeling hierarchical user novelty-seeking intent in recommendations through extensive experiments on simulated and real-world datasets. In particular, we demonstrate that the effectiveness of our proposed hierarchical RL-based method lies in its ability to capture such hierarchically-structured intent. As a result, the proposed HRL model achieves superior performance on several public datasets, compared with state-of-art baselines.

Spoken language recognition (SLR) is the task of automatically identifying the language present in a speech signal. Existing SLR models are either too computationally expensive or too large to run effectively on devices with limited resources. For real-world deployment, a model should also gracefully handle unseen languages outside of the target language set, yet prior work has focused on closed-set classification where all input languages are known a-priori. In this paper we address these two limitations: we explore efficient model architectures for SLR based on convolutional networks, and propose a multilabel training strategy to handle non-target languages at inference time. Using the VoxLingua107 dataset, we show that our models obtain competitive results while being orders of magnitude smaller and faster than current state-of-the-art methods, and that our multilabel strategy is more robust to unseen non-target languages compared to multiclass classification.