Product assortment selection is a critical challenge facing physical retailers. Effectively aligning inventory with the preferences of shoppers can increase sales and decrease out-of-stocks. However, in real-world settings the problem is challenging due to the combinatorial explosion of product assortment possibilities. Consumer preferences are typically heterogeneous across space and time, making inventory-preference alignment challenging. Additionally, existing strategies rely on syndicated data, which tends to be aggregated, low resolution, and suffer from high latency. To solve these challenges, we introduce a real-time recommendation system, which we call EdgeRec3D. Our system utilizes recent advances in 3D computer vision for perception and automatic, fine grained sales estimation. These perceptual components run on the edge of the network and facilitate real-time reward signals. Additionally, we develop a Bayesian payoff model to account for noisy estimates from 3D LIDAR data. We rely on spatial clustering to allow the system to adapt to heterogeneous consumer preferences, and a graph-based candidate generation algorithm to address the combinatorial search problem. We test our system in real-world stores across two, 6-8 week A/B tests with beverage products and demonstrate a 35% and 27% increase in sales respectively. Finally, we monitor the deployed system for a period of 28 weeks with an observational study and show a 9.4% increase in sales.

Current recommendation systems are significantly affected by a serious issue of temporal data shift, which is the inconsistency between the distribution of historical data and that of online data. Most existing models focus on utilizing updated data, overlooking the transferable, temporal data shift-free information that can be learned from shifting data. We propose the Temporal Invariance of Association theorem, which suggests that given a fixed search space, the relationship between the data and the data in the search space keeps invariant over time. Leveraging this principle, we designed a retrieval-based recommendation system framework that can train a data shift-free relevance network using shifting data, significantly enhancing the predictive performance of the original model in the recommendation system. However, retrieval-based recommendation models face substantial inference time costs when deployed online. To address this, we further designed a distill framework that can distill information from the relevance network into a parameterized module using shifting data. The distilled model can be deployed online alongside the original model, with only a minimal increase in inference time. Extensive experiments on multiple real datasets demonstrate that our framework significantly improves the performance of the original model by utilizing shifting data.

Recommendation Systems (RS) are often plagued by popularity bias. When training a recommendation model on a typically long-tailed dataset, the model tends to not only inherit this bias but often exacerbate it, resulting in over-representation of popular items in the recommendation lists. This study conducts comprehensive empirical and theoretical analyses to expose the root causes of this phenomenon, yielding two core insights: 1) Item popularity is memorized in the principal spectrum of the score matrix predicted by the recommendation model; 2) The dimension collapse phenomenon amplifies the relative prominence of the principal spectrum, thereby intensifying the popularity bias. Building on these insights, we propose a novel debiasing strategy that leverages a spectral norm regularizer to penalize the magnitude of the principal singular value. We have developed an efficient algorithm to expedite the calculation of the spectral norm by exploiting the spectral property of the score matrix. Extensive experiments across seven real-world datasets and three testing paradigms have been conducted to validate the superiority of the proposed method.

Along with the development of chatbot, language models and speech technologies, there is a growing possibility and interest of creating systems able to interface with humans seamlessly through natural language or directly via speech. In this paper, we want to demonstrate that placing the research on dialog system in the broader context of embodied intelligence allows to introduce concepts taken from neurobiology and neuropsychology to define behavior architecture that reconcile hand-crafted design and artificial neural network and open the gate to future new learning approaches like imitation or learning by instruction. To do so, this paper presents a neural behavior engine that allows creation of mixed initiative dialog and action generation based on hand-crafted models using a graphical language. A demonstration of the usability of such brain-like inspired architecture together with a graphical dialog model is described through a virtual receptionist application running on a semi-public space.

In rebranding artificial intelligence as Apple Intelligence, Apple Inc is banking on the idea that people by and large won't buy the powerful A.I. software that its rivals are developing. Instead, they'll want really cool hardware that incorporates A.I. It's a compelling but risky strategy for a company that specializes in seamlessly integrating hardware and software into must-have products. "It's the next big step for Apple," Apple CEO Tim Cook said Monday in unveiling Apple Intelligence at the company's developers conference. Apple is diving into artificial intelligence – focused on the idea of a "virtual personal assistant" - as a potential must-have app for consumers. Since it lacks its own cutting-edge version of the predictive, sounds-like-a-human technology known as generative A.I., Apple will license that technology from other companies, starting with OpenAI.

Large language model (LLM)-based mobile agents are increasingly popular due to their capability to interact directly with mobile phone Graphic User Interfaces (GUIs) and their potential to autonomously manage daily tasks. Despite their promising prospects in both academic and industrial sectors, little research has focused on benchmarking the performance of existing mobile agents, due to the inexhaustible states of apps and the vague definition of feasible action sequences. To address this challenge, we propose an efficient and user-friendly benchmark, MobileAgentBench, designed to alleviate the burden of extensive manual testing. We initially define 100 tasks across 10 open-source apps, categorized by multiple levels of difficulty. Subsequently, we evaluate several existing mobile agents, including AppAgent and MobileAgent, to thoroughly and systematically compare their performance. All materials are accessible on our project webpage: https://MobileAgentBench.github.io,

The pitch offers an implicit contrast with the likes of Alphabet, Amazon, or Meta, which collect and store enormous amounts of personal data. Apple says any personal data passed on to the cloud will be used only for the AI task at hand and will not be retained or accessible to the company, even for debugging or quality control, after the model completes the request. Simply put, Apple is saying people can trust it to analyze incredibly sensitive data--photos, messages, and emails that contain intimate details of our lives--and deliver automated services based on what it finds there, without actually storing the data online or making any of it vulnerable. It showed a few examples of how this will work in upcoming versions of iOS. Instead of scrolling through your messages for that podcast your friend sent you, for example, you could simply ask Siri to find and play it for you.

Apple Intelligence is coming, but not to every iPhone out there. In fact, you'll need to have a device with an A17 Pro processor or M-series chip to use many of the features unveiled during the Apple Intelligence portion of WWDC 2024. That means only iPhone 15 Pro owners (and those with an M-series iPad) will get the iOS 18-related Apple Intelligence (AI?) updates like Genmoji, Image Playground, the redesigned Siri and Writing Tools. It's not evident exactly why older devices using an A16 chip (like the iPhone 14 Pro) won't work with Apple Intelligence, given its neural engine seems more than capable compared to the M1. A closer look at the specs sheets of those two processors show that the main differences appear to be in memory and GPU prowess.

In addition to revealing its initial plans for AI and annual updates to iOS, macOS and more at WWDC 2024, Apple also discussed new capabilities coming to the second-gen AirPods Pro. Siri Interactions will allow you to respond to the assistant by nodding your head yes or shaking your head no. Apple also plans to introduce improved Voice Isolation that further reduces background noise when you're on a call. Both of these items are exclusive to the most recent AirPods Pro, because they rely on the company's H2 chip like existing Adaptive Audio, Personalized Volume and Conversation Awareness features. Like those advanced audio tools that are already available on AirPods Pro, Siri Interactions and Voice Isolation use the processing abilities of the H2 chip in tandem with the power of a source device -- an iPhone or MacBook Pro, for example.

One of the bigger selling points of Apple's AirPods for some people is their unsealed design, which means they rest just outside of your ear canal instead of inserting all the way in. Open-style earbuds like these aren't good at blocking out ambient noise as a result, but they tend to be more comfortable to wear for those with sensitive ears. If this idea sounds appealing but you're on a tighter budget, the latest Amazon Echo Buds are a similar alternative that we recommend in our guide to the best budget earbuds. They normally cost 50, but a new deal at Amazon has dropped them back down to 35. That matches the lowest price we've tracked.