While other areas of machine learning have seen more and more automation, designing a high-performing recommender system still requires a high level of human effort. Furthermore, recent work has shown that modern recommender system algorithms do not always improve over well-tuned baselines. A natural follow-up question is, "how do we choose the right algorithm for a new dataset and performance metric?" In this work, we start by giving the first large-scale study of recommender system approaches by comparing 24 algorithms and 100 sets of hyperparameters across 85 datasets and 315 metrics. We find that the best algorithms and hyperparameters are highly dependent on the dataset and performance metric. However, there is also a strong correlation between the performance of each algorithm and various meta-features of the datasets. Motivated by these findings, we create RecZilla, a meta-learning approach to recommender systems that uses a model to predict the best algorithm and hyperparameters for new, unseen datasets. By using far more meta-training data than prior work, RecZilla is able to substantially reduce the level of human involvement when faced with a new recommender system application.

Voice commands, typically accessed through mobile-banking applications, can now be used to complete a vast array of banking transactions, including sending payments, processing foreign-exchange transactions, logging in and out of accounts, and confirming purchases. All of these tasks are made more efficient through the use of speech. But today's voice-banking solutions are not only being designed to complete simple tasks. They can also provide answers to questions in the same vein as Siri or Alexa, and they also seek to integrate other tech such as artificial intelligence (AI) to more accurately gauge customer behaviour. This opens up a myriad of additional, and often more complex, actionable options for the customer.

NLP is not something all data scientists necessarily work with and are required to know. Whether or not you are, depends on the company interviewing you for a data science position. Well, you'll have to know what it is so you can avoid it in your career, if nothing else. In case you're intrigued by NLP and willing to learn more, you will benefit from knowing what interview questions you could expect. No, it's not that pseudoscientific psychological approach that gained popularity recently.

We study embedding table placement for distributed recommender systems, which aims to partition and place the tables on multiple hardware devices (e.g., GPUs) to balance the computation and communication costs. Although prior work has explored learning-based approaches for the device placement of computational graphs, embedding table placement remains to be a challenging problem because of 1) the operation fusion of embedding tables, and 2) the generalizability requirement on unseen placement tasks with different numbers of tables and/or devices. To this end, we present DreamShard, a reinforcement learning (RL) approach for embedding table placement. DreamShard achieves the reasoning of operation fusion and generalizability with 1) a cost network to directly predict the costs of the fused operation, and 2) a policy network that is efficiently trained on an estimated Markov decision process (MDP) without real GPU execution, where the states and the rewards are estimated with the cost network. Equipped with sum and max representation reductions, the two networks can directly generalize to any unseen tasks with different numbers of tables and/or devices without fine-tuning. Extensive experiments show that DreamShard substantially outperforms the existing human expert and RNN-based strategies with up to 19% speedup over the strongest baseline on large-scale synthetic tables and our production tables. The code is available at https://github.com/daochenzha/dreamshard

As virtual personal assistants have now penetrated the consumer market, with products such as Siri and Alexa, the research community has produced several works on task-oriented dialogue tasks such as hotel booking, restaurant booking, and movie recommendation. Assisting users to cook is one of these tasks that are expected to be solved by intelligent assistants, where ingredients and their corresponding attributes, such as name, unit, and quantity, should be provided to users precisely and promptly. However, existing ingredient information scraped from the cooking website is in the unstructured form with huge variation in the lexical structure, for example, "1 garlic clove, crushed", and "1 (8 ounce) package cream cheese, softened", making it difficult to extract information exactly. To provide an engaged and successful conversational service to users for cooking tasks, we propose a new ingredient parsing model that can parse an ingredient phrase of recipes into the structure form with its corresponding attributes with over 0.93 F1-score. Experimental results show that our model achieves state-of-the-art performance on AllRecipes and Food.com datasets.

Google will be taking the wraps off its Pixel 7 smartphone and Pixel Watch smartwatch at its Made by Google event in New York City later this week, but today, Google-owned Nest is sharing a few new products and updates in the company's smart-home ecosystem--from a new Nest Wifi Pro router with Wi-Fi 6E support to a redesigned Google Home app. Last year, Nest debuted a second-generation Nest Doorbell (Battery), a battery-powered video doorbell. Now, it's time for a second-gen wired version, for those who don't want to worry about their doorbell running out of juice. It looks quite similar and has similar specs but is 30 percent smaller. There's 24/7 recording support, and it stores three hours of important events in its local memory in case your Wi-Fi goes out. The new doorbell's camera isn't as high-resolution as the original Nest Doorbell, with a 960 x 1,280-pixel resolution, but it's the HDR support that takes the camera quality a step further--it'll be able to handle bright lights and better expose your footage.

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 dropped unmissable discounts on their bestselling Echo, Blink and Fire devices ahead of their Prime Early Access Sale. Amazon has announced it will run its first-ever Prime Early Access Sale ahead of Black Friday, spanning 48 hours from 11 October through 12 October. The Prime Early Access Sale will offer hundreds of thousands of deals, but bargain hunters can already enjoy impressive savings on Amazon Echo and Alexa devices - and you don't have to be a Prime member.

In recent years, between lockdowns, curfews, supply chain disruptions, and energy crunches, retailers must have felt like dinosaurs trying to dodge a rain of asteroids and avoid extinction. But unlike those giant prehistoric reptiles, the retail industry could count on a full array of technological innovations to better meet the challenges of these difficult times. One of the most impactful tools in this arsenal has certainly turned out to be artificial intelligence, including its powerful sub-branch known as machine learning (ML). Let's briefly frame the nature of this technology and explore the key use cases of machine learning in retail. Machine learning in retail relies on self-improving computer algorithms created to process data, spot recurring patterns and anomalies among variables, and autonomously learn how such relations affect or determine the industry's trends, phenomena, and business scenarios.

Context has been an important topic in recommender systems over the past two decades. A standard representational approach to context assumes that contextual variables and their structures are known in an application. Most of the prior CARS papers following representational approach manually selected and considered only a few crucial contextual variables in an application, such as time, location, and company of a person. This prior work demonstrated significant recommendation performance improvements when various CARS-based methods have been deployed in numerous applications. However, some recommender systems applications deal with a much bigger and broader types of contexts, and manually identifying and capturing a few contextual variables is not sufficient in such cases. In this paper, we study such ``context-rich'' applications dealing with a large variety of different types of contexts. We demonstrate that supporting only a few most important contextual variables, although useful, is not sufficient. In our study, we focus on the application that recommends various banking products to commercial customers within the context of dialogues initiated by customer service representatives. In this application, we managed to identify over two hundred types of contextual variables. Sorting those variables by their importance forms the Long Tail of Context (LTC). In this paper, we empirically demonstrate that LTC matters and using all these contextual variables from the Long Tail leads to significant improvements in recommendation performance.

Artificial intelligence, also known as AI, involves using computers to carry out tasks that would normally require human intelligence, such as understanding natural language and recognizing patterns. AI works by using a number of different methods to mimic human intelligence, including machine learning, natural language processing and artificial neural networks. An artificial intelligence contact center is a customer service center that uses AI technology to help employees respond to customer inquiries. It can be used to automate tasks such as routing calls, providing information about products and services, and handling complaints. What are chatbots and virtual assistants?