All the sessions from Transform 2021 are available on-demand now. Hungryroot, an AI-powered delivery service, hopes to occupy a similar niche for online groceries in the United States. The recommender system uses a collaborative filtering, supervised learning model to match consumer preferences to foods. Customers answer questions about their dietary habits, the kinds of foods they (and family members) like, the family size, budget, and more. On a weekly basis, the Hungryroot algorithm predicts the groceries the customer might like.

So far, most research on recommender systems focused on maintaining long-term user engagement and satisfaction, by promoting relevant and personalized content. However, it is still very challenging to evaluate the quality and the reliability of this content. In this paper, we propose FEBR (Expert-Based Recommendation Framework), an apprenticeship learning framework to assess the quality of the recommended content on online platforms. The framework exploits the demonstrated trajectories of an expert (assumed to be reliable) in a recommendation evaluation environment, to recover an unknown utility function. This function is used to learn an optimal policy describing the expert's behavior, which is then used in the framework to provide high-quality and personalized recommendations. We evaluate the performance of our solution through a user interest simulation environment (using RecSim). We simulate interactions under the aforementioned expert policy for videos recommendation, and compare its efficiency with standard recommendation methods. The results show that our approach provides a significant gain in terms of content quality, evaluated by experts and watched by users, while maintaining almost the same watch time as the baseline approaches.

Every year, nearly 20 million students enroll in universities across the United States. Given how important the choice of a college can be for a person's academic and professional career, students understandably have a lotof questions about majors, eligibility, tuition, campus life, faculty, and more, while going through the process of selecting a university. And as universities compete to get on board students who are the right'fit', it is important for them to not only provide the right answers to prospective students' queries but also to effectively communicate precisely what they have to offer. Colleges invest a lot of time and resources in engaging with prospective students -- with US universities spending billions of dollars annually on admissions. But the COVID-19 pandemic has thrown open new challenges for universities, as they now have to find ways to engage with students at a time when in-person interactions, including visits to university campuses and college fairs, will likely not be possible.

Those who find Alexa's default voice too cold can have the digital assistant mimic celebs. Though, in the past, the only A-list impression the AI could do was everyone's fave badass Samuel L. Jackson. Turns out, people liked the idea of ordering a weather report from a superstar, because Amazon is adding two more famous voices to Alexa's toolkit. The new options include four-time NBA champ Shaquille O'Neal and Oscar-nominated actor Melissa McCarthy. Amazon says the SLJ skill, introduced for a limited price of $0.99, became one of its top-selling digital purchases upon launch.

First unveiled in 2018 at CES, LG is finally bringing its futuristic, rollable LG Signature OLED R 65-inch 4K TV to the US for the hefty sum of $100,000. According to LG's website, the television has an OLED screen, a Dolby Atmos and Sound Pro sound system and has both Google Assistant and Amazon's Alexa for built-in voice control. The 65-inch screen, which is described as a'revolutionary new experience,' gets tucked into aluminum housing unit when not in use, either for watching movies or playing video games. It has an OLED screen, a Dolby Atmos and Sound Pro sound system and both Google Assistant and Amazon's Alexa for built-in voice control The cabinet also has a Dolby Atmos sound system of its own. Manufactured in LG's Gumi facility, each TV is painstakingly assembled'with craftsman-like skill with attention to every detail', LG said last year.

It has been over a year since international travel as we knew it ground to a halt. When the COVID-19 pandemic hit, air travel in the US dropped by 95% – from around two million travellers per day to fewer than 100,000. Until recently, flights in and out of Australia have been limited to those trying to get home, reuniting with their loved ones, or fleeing places that were no longer safe. Slowly, vaccination is making the possibility of taking to the skies again seem within reach. But what might have changed?

Microsoft is planning to bring Clippy, its Office 97 assistant that looks like a paperclip, back to Windows. The software giant tweeted a photo of Clippy saying that, if the post received 20,000 likes, it would replace the paperclip emoji in Microsoft Office with Clippy. At time of writing, the tweet has nearly 134,000 likes. Clippy infamously offered support in Microsoft Word and other programs, such as Microsoft Publisher, but by 2008 the assistant was killed. Windows 10, Microsoft's current operating system, uses Cortana as its virtual assistant – a name taken from the Halo game franchise.

Fox News Flash top headlines are here. Check out what's clicking on Foxnews.com. A woman was allegedly held captive in a California home for three days by a man she met on Tinder. "On July 12th, 2021 at approximately 16:59, Oakland Police Officers were dispatched to the 5400 Block of Fleming Avenue to investigate a report of a kidnapping," the Oakland Police Department said in a statement of the incident. "A preliminary investigation revealed that an adult female (Non-Oakland resident) was falsely imprisoned and sexually assaulted by her male partner."

Sony's latest top-of-the-range noise-cancelling earbuds are a cut above the previous generation and the competition. They are the successors to the WF-1000XM3, which were great but huge. Sony has significantly shrunk the case and earbuds for the mark four model without sacrificing performance. The earbuds are fairly heavy, weighing 7.3g each, compared with the 5.4g AirPods Pro, but they don't feel so in the ear, and they are comfortable for hours of listening without a break. They don't have stalks but protrude further than some slimmer rivals.

Sequential recommender systems (SRS) have become a research hotspot due to its power in modeling user dynamic interests and sequential behavioral patterns. To maximize model expressive ability, a default choice is to apply a larger and deeper network architecture, which, however, often brings high network latency when generating online recommendations. Naturally, we argue that compressing the heavy recommendation models into middle- or light- weight neural networks is of great importance for practical production systems. To realize such a goal, we propose AdaRec, a knowledge distillation (KD) framework which compresses knowledge of a teacher model into a student model adaptively according to its recommendation scene by using differentiable Neural Architecture Search (NAS). Specifically, we introduce a target-oriented distillation loss to guide the structure search process for finding the student network architecture, and a cost-sensitive loss as constraints for model size, which achieves a superior trade-off between recommendation effectiveness and efficiency. In addition, we leverage Earth Mover's Distance (EMD) to realize many-to-many layer mapping during knowledge distillation, which enables each intermediate student layer to learn from other intermediate teacher layers adaptively. Extensive experiments on real-world recommendation datasets demonstrate that our model achieves competitive or better accuracy with notable inference speedup comparing to strong counterparts, while discovering diverse neural architectures for sequential recommender models under different recommendation scenes.