The word "deep" in "deep learning" refers to the number of layers through which the data is transformed. Deep models are able to extract better features than shallow models and hence, extra layers help in learning the features effectively.

Recommendation is a prevalent application of machine learning that affects many users; therefore, it is important for recommender models to be accurate and interpretable. In this work, we propose a method to both interpret and augment the predictions of black-box recommender systems. In particular, we propose to interpret feature interactions from a source recommender model and explicitly encode these interactions in a target recommender model, where both source and target models are black-boxes. By not assuming the structure of the recommender system, our approach can be used in general settings. In our experiments, we focus on a prominent use of machine learning recommendation: ad-click prediction. We found that our interaction interpretations are both informative and predictive, e.g., significantly outperforming existing recommender models. What's more, the same approach to interpret interactions can provide new insights into domains even beyond recommendation, such as text and image classification.

Effective optimization is essential for real-world interactive systems to provide a satisfactory user experience in response to changing user behavior. However, it is often challenging to find an objective to optimize for interactive systems (e.g., policy learning in task-oriented dialog systems). Generally, such objectives are manually crafted and rarely capture complex user needs in an accurate manner. We propose an approach that infers the objective directly from observed user interactions. These inferences can be made regardless of prior knowledge and across different types of user behavior. We introduce Interactive System Optimizer (ISO), a novel algorithm that uses these inferred objectives for optimization. Our main contribution is a new general principled approach to optimizing interactive systems using data-driven objectives. We demonstrate the high effectiveness of ISO over several simulations.

The upcoming iOS 14 software release will include some notable changes to the Apple's Podcasts app, such as a feature that suggests podcasts you may like based on what you currently listen to, according to 9to5Mac. The leak comes ahead of the virtual WWDC next Monday, where Apple is expected to reveal more about iOS 14. The new podcast feature will reportedly work similar to Apple Music's "For You" feature, which uses an algorithm to suggest new songs, albums and artists, a source told 9to5Mac. Apple Podcasts will learn from your listening preferences and suggest new podcasts. Another reported update to Apple Podcasts is the option for podcast creators to offer bonus content -- similar to what is currently offered when you buy a movie Apple TV.

As protests continue both in the United States and the United Kingdom over police brutality and the death of George Floyd, an iPhone shortcut that allows users to quickly record interactions with law enforcement is becoming more popular. The tool, which was created in 2018, uses Apple's Shortcuts app. This allows users to trigger a series of actions based on a single word, similar to Routines with Amazon's Alexa voice assistant. Saying "Hey Siri, I'm getting pulled over" will automatically activate the program, which pauses any music that could be playing, turns down the brightness on the phone screen, and puts the device on Do Not Disturb. It can also send a contact a text message, letting them know that you have been pulled over by the police as well as sending your exact location.

Amazon has released its first Echo device for use outside of the house, allowing users to take Alexa in their car. The company revealed the device in 2018 but it has finally come to customers in the UK and Ireland. Echo Auto plugs into a car's 12V power outlet or built-in USB port and connects to the in-car stereo via either audio jack cable or Bluetooth to enable the use of voice assistant Alexa inside the vehicle. Users are then able to use Alexa voice commands to control music, check the news, make phone calls or check their schedule without taking hands off the wheel or eyes off the road. The device gets internet connectivity by connecting to a user's smartphone and the Alexa app and using its existing data plan.

In this paper we discuss approaches to evaluating and validating the ethical claims of a Conversational AI system. We outline considerations around both a top-down regulatory approach and bottom-up processes. We describe the ethical basis for each approach and propose a hybrid which we demonstrate by taking the case of a customer service chatbot as an example. We speculate on the kinds of top-down and bottom-up processes that would need to exist for a hybrid framework to successfully function as both an enabler as well as a shepherd among multiple use-cases and multiple competing AI solutions.

Most of existing embedding based recommendation models use embeddings (vectors) corresponding to a single fixed point in low-dimensional space, to represent users and items. Such embeddings fail to precisely represent the users/items with uncertainty often observed in recommender systems. Addressing this problem, we propose a unified deep recommendation framework employing Gaussian embeddings, which are proven adaptive to uncertain preferences exhibited by some users, resulting in better user representations and recommendation performance. Furthermore, our framework adopts Monte-Carlo sampling and convolutional neural networks to compute the correlation between the objective user and the candidate item, based on which precise recommendations are achieved. Our extensive experiments on two benchmark datasets not only justify that our proposed Gaussian embeddings capture the uncertainty of users very well, but also demonstrate its superior performance over the state-of-the-art recommendation models.

Google wants to make the Assistant more powerful, and is announcing today a slew of developer-centric updates that will make it easier to build tools for the digital helper. Three new features in particular are most intriguing: Home Storage, Continuous Match mode and AMP support on smart displays. Google is also releasing a new web-based console and SDK that will make it easier for developers to create for Assistant. These announcements are the largest since the company acquired conversational user experience platform Dialogflow in 2016, according to director of product management Payam Shodjai. With Home Storage, programmers can tap into a pool of information shared by devices in the same household (or, in developer speak, connected on the home graph).

Better yet, all of these gadgets can be delivered next day to your door, which is perfect if you're looking for a last-minute gift. Savings include a mega discount on Amazon's most popular Alexa smart speaker, which has been slashed to its lowest price to date this year. Plus, you can pre-order the all-new Fire 7 Tablet now and receive 17 per cent off. What dad wouldn't appreciate new tech this Father's Day? Ahead, check out the best Amazon device deals that would make great gifts. If your dad is after a smart speaker, then this deal is for you.