pr23579

Amanda Stent

When you think of a computer scientist, what comes to mind? Perhaps someone in a lab coat or a person who has been working in their basement for several days straight. Either way, this perception isn’t drawing young girls to enter the field of computer science.

That’s the image that AT&T Labs researcher Amanda Stent has set out to reverse. Amanda co-authored the book “The Princess at the Keyboard: Why Girls Should Become Computer Scientists.” Not only does Amanda’s book speak to her dedication of engaging more girls to be computer scientists, but her work at AT&T Labs Research is a daily testament to successful women in STEM.

Amanda’s work revolves around language processing, spoken dialogue systems and interactive systems that use language. She has used her education and experience in the field to work on projects such as the U-verse Easy Remote app, virtual assistants and more. Amanda explains, “I am really interested in how people with disabilities can use our mobile phones, television and so on to make their lives easier and give them more independence.”

We recently sat down with Amanda to talk more about her background, work and how she views innovation. Below is an excerpt from that discussion.

Innovation is doing something new with building blocks that you already have to hand. And beyond that is invention, which is doing something new with building blocks that you may have to create.

What inspired you to go into computer science?
I have really liked computer science since I was in high school, during which I took a class in artificial intelligence. There was this programming language they taught us called Prolog – a logic-based programming language. So, you write down what you think will be true about the world. Then you write down rules that then allow you to decide what other things will be true about the world, to reason, and make new things that are true about the world. I just thought that the programming language was so beautiful, I fell in love with it. I wanted to keep doing that sort of thing for the rest of my life, and that’s how I got into computer science.

Tell me about your background and how you found your way to AT&T Labs.
I have a Ph.D. in computer science. Previously, I was a faculty member at a university for several years, but I was collaborating with people here at AT&T and, in particular, my area of research – language and dialogue systems. There are some really terrific people here at AT&T, and I just like working with them so much that I moved here.

Tell me some of the projects you’re working on.
One project that I am working on – along with a bunch of other people in Labs Research – is this whole area of virtual assistants. That is, programs that live in the network that can help you with your daily life tasks. They could be maintaining your calendar, sending e-mails, or going shopping, which people do a lot every day. They could also be making hotel reservations or tickets, which people do a little less frequently.

What we are doing here is trying to create systems that you can interact with over your mobile phone, desktop, or TV using speech, gesture and the way your eyes are looking to make it easy for people to get help from the computers doing these tasks.

AT&T’s innovations are sometimes invisible to a lot of people, but they really underlie almost everything. Our network, the programs that run in our network, the programming languages that run the computer – a lot of those things were developed and came out of AT&T. You just don’t see them so much, but they make everything that you do in your daily life possible.

How would you say that your work affects the consumer?
Assistive technology means you’re making technology more accessible for everybody – the young, the old, people with disabilities. As a company, we are uniquely positioned to help make people’s lives really easy, give them a tremendous amount of independence and a tremendous amount of privacy. I am really interested in how people with disabilities can use our mobile phones, television and so on to make their lives easier and give them more independence.

An example of that is AT&T just released an accessible TV remote {U-verse Easy Remote} in the App Store. So, if you want to look for a show instead of watching all the channels scroll by, you can just say the show you want to watch – or the actor or movie – and it will read you them one-by-one and you can choose the one you want by pressing a button. It’s one-button interaction with the TV guide.

How would you define innovation?
Innovation is doing something new with building blocks that you already have to hand. And beyond that is invention, which is doing something new with building blocks that you may have to create. Here in Labs Research, we do some innovation and some invention.

How do you think AT&T pushes the envelope in terms of innovation?
What I say to people who ask me about AT&T is that a lot of companies’ innovations are right in your face. You go to their webpage, and there it is; or, you get a phone, and you are holding it in your hand. AT&T’s innovations are sometimes invisible to a lot of people, but they really underlie almost everything. Our network, the programs that run in our network, the programming languages that run the computer – a lot of those things were developed and came out of AT&T. You just don’t see them so much, but they make everything that you do in your daily life possible.

How can a computer help you decide which stores to go to that will optimize your time and minimize your costs? This is a really hard problem we were trying to solve in the area of personal assistant capbilities.

When we were thinking about I, I suddenly thought to myself, this is exactly like another problem in a completely different area that I work on, which is – how do you know when someone is talking that the phrases they use to refer to people and things like “he, she, it, the boy, the dog” refer to the same things or different things? You are not going to think that these problems are related but they actually are.

In each case, it’s finding a route that optimizes function; it turns out, a problem called the “Traveling Purchaser Problem,” which is a relatively long problem in computer science and operations research. Once we had that figured out, we knew how to implement it.