Artificial intelligence (AI) is becoming a bigger part of our lives in several ways, and we interact with it in different forms almost every day. From booking a cab on a virtual assistant to interacting with a customer service agent, AI works in the background to help make the experience better. And AI will proliferate even faster in the coming years. JPMorgan, for instance, believes that sales of AI-related software, hardware, and other services will grow from an estimated $12 billion last year to roughly $58 billion in 2021. There are several ways for investors to tap into the AI trend.

From neural networks and machine learning automation to virtual assistants and human-like robots, artificial intelligence (AI) encompasses many different things. But when AI is envisioned, many people automatically think of pure AI -- software that has equal or superior cognitive capabilities to the humans who created it. This can conjure images of an android butler announcing "dinner is served" or a robot-like Rosie from the Jetsons cleaning the house or helping Junior with his homework. For other people, pure AI can conjure frightening images of a future society controlled by Terminator-like robots that have too much knowledge and power. As with all unknowns, the future vision of an AI-driven world can be both exciting and scary, yet it is a world that we're heading toward nevertheless.

Imagine it's the early 1990s and you've heard about a new platform called "the internet." You realize its potential for customer communication, so you want to build a web presence. You want to create a site that reflects your brand standards and communicates with customers in your brand's voice. That's exactly what some brands are doing today with AI assistants. Sales of voice-driven AI assistants like Amazon's Echo and Google Home are booming, and it's increasingly common for consumers to ask Alexa or Google about products or services they're interested in instead of entering keywords into a search engine.

As you may already be aware, Google Duplex is a new technology launched by Google at Google I/O 2018, the company's annual developer conference. Google Duplex is an AI system that is designed to accomplish real-world tasks over the phone, by mimicking human conversation. Google CEO Sundar Pichai personally demonstrated phones calls made by this digital voice assistant. The product was demoed by showing a phone call where the AI assistant makes an appointment at the hairdresser's, by calling them and having an actual conversation. Now chatbots and voice assistants have been around for a while now, so that probably doesn't sound all that intriguing to you.

The Google Assistant is getting better and better and that's not going to change any time soon, especially considering Google's artificial intelligence prowess. But perhaps the best thing about the Google Assistant has nothing to do with the Assistant itself, but more to do with the fact that it can be used to control such a huge range of smart home devices.

NEW YORK – Alexa for couch potatoes is coming: Amazon's new streaming TV device will let users shout out when they want to turn on the TV, flip channels or search for sitcoms -- all without pushing any buttons. The Fire TV Cube is not entirely hands-free yet, however. Some apps or streaming services may require viewers to pick up the included remote to rewind or stop a show. Inc. says it's working with those services to integrate voice commands. Amazon's other voice-controlled Fire TV devices require a push of the remote's mic button or a separate Echo device with Amazon's Alexa voice assistant.

In terms of computer applications, we will see increasing application and adoption of machine learning (ML) and artificial intelligence (AI) techniques. We already see this in shopping recommendations, games, and large social networks. Voice assistants such as Siri, Alexa, and Google Assistant use ML to perform natural language processing and classification to respond appropriately. Such techniques will make interacting with devices increasingly seamless, which will ultimately make technology easier to use while making humans more efficient in finding and managing information. Just like automation and the application of machine learning to businesses and business processes, there are many opportunities to apply similar techniques to government functions.

If you had a real-life personal assistant, he could help you stay on task by reminding you about the tasks you need to do and keeping you on schedule. A mobile AI does the same thing by making it much easier to set reminders, alarms and timers. To make sure you don't forget an item on your to-do list, have your AI assistant remind you about it. Tell your phone, "Remind me to..." or "Set a reminder for..." to create the note. Then decide whether you'd like to receive the reminder at a specific date and time, or when you travel to a specific location.

Nowadays, artificial intelligence (AI) is being used in a variety of ways, the most common of which is in improving computer programs and technologies. You might have heard of the popular artificial intelligence conversation program SimSimi back in 2012. For those who don't know this piece of information, SimSimi was created by ISMaker, and it has an application designed for Android, for Windows Phone, and for iOS. Another artificial intelligence application is in application programs that function as AI assistants, such as Alexa and Siri. These AI assistants help businessmen run their businesses more efficiently.

This paper explores the adaptively (active) PAC (probably approximately correct) top-$k$ ranking and total ranking from $l$-wise ($l\geq 2$) comparisons under the popular multinomial logit (MNL) model. By adaptively choosing sets to query and observing the noisy output about the most favored item of each query, we want to design ranking algorithms that recover the top-$k$ or total ranking using as few queries as possible. For the PAC top-$k$ ranking problem, we prove a lower bound on the sample complexity (aka number of queries), and propose an algorithm that is sample complexity optimal up to a $O(\log(k+l)/\log{k})$ factor. When $l=2$ (i.e., pairwise) or $l=O(poly(k))$, the algorithm matches the lower bound. For the PAC total ranking problem, we prove a lower bound, and propose an algorithm that matches the lower bound. When $l=2$, this model reduces to the popular Plackett-Luce (PL) model, and our results still outperform the state-of-the-art theoretically and numerically. We also run comparisons of our algorithms with the state-of-the-art on synthesized data as well as real-world data, and demonstrate the improvement on sample complexity numerically.