back to indexRohit Prasad: Alexa Prize | AI Podcast Clips
00:00:10.120 |
and what have you learned and what's surprising? 00:00:15.920 |
- Absolutely, it's a very exciting competition. 00:00:24.340 |
where we threw the gauntlet to the universities 00:00:29.400 |
to say, can you build what we call a social bot 00:00:39.960 |
talking to someone who you're meeting for the first time 00:00:55.200 |
We have completed two successful years of the competition. 00:00:59.080 |
The first was one with the University of Washington, 00:01:04.360 |
We have an extremely strong team of 10 cohorts 00:01:07.100 |
and the third instance of the Alexa Prize is underway now. 00:01:29.040 |
- Just a few quick questions, sorry for the interruption. 00:01:31.360 |
What does failure look like in the 20 minute session? 00:01:44.020 |
but the quality of the conversation too that matters. 00:01:48.960 |
before I answer that question on what failure means 00:02:07.880 |
and we have interactors who interact with these social bots, 00:02:16.440 |
all the judging is essentially by the customers of Alexa. 00:02:20.200 |
And there you basically rate on a simple question 00:02:25.920 |
So that's where we are not testing for a 20 minute 00:02:30.280 |
because you do want it to be very much like a 00:02:37.560 |
So did you really break that 20 minute barrier 00:02:40.280 |
is why we have to test it in a more controlled setting 00:02:54.480 |
with real customers versus in the lab to award the prize. 00:03:05.440 |
and two of them have to say this conversation 00:03:18.280 |
How far, so the DARPA challenge in the first year, 00:03:35.120 |
to the extent that we're definitely not close 00:03:37.840 |
to the 20 minute barrier being with coherence 00:03:51.480 |
in what kind of responses these social bots generate 00:03:56.600 |
What's even amazing to see that now there's humor coming in. 00:04:04.000 |
- You're talking about ultimate science of intelligence. 00:04:22.440 |
not only what we think of natural language abilities, 00:04:27.800 |
and aspects of when to inject an appropriate joke, 00:04:35.840 |
how you come back with something more intelligible 00:04:42.600 |
and we are domain experts, we can speak to it. 00:04:44.920 |
But if you suddenly switch the topic to that, 00:04:46.720 |
I don't know of, how do I change the conversation? 00:04:49.560 |
So you're starting to notice these elements as well. 00:05:03.000 |
and essentially mask some of the understanding defects 00:05:07.280 |
- So some of this, this is not Alexa the product. 00:05:15.240 |
I have a question sort of in this modern era, 00:05:24.560 |
And some things that are a little bit too edgy, 00:05:29.720 |
are people in this context pushing the limits? 00:05:43.400 |
as part of the dialogue to really draw people in? 00:05:51.760 |
I think fun is more part of the engaging part for customers. 00:06:01.840 |
But that apart, the real goal was essentially 00:06:04.680 |
what was happening is with a lot of AI research 00:06:09.360 |
we felt that academia has the risk of not being able 00:06:12.480 |
to have the same resources at disposal that we have, 00:06:15.640 |
which is lots of data, massive computing power, 00:06:26.000 |
So we brought all these three together in the Alexa Prize. 00:06:28.360 |
That's why it's one of my favorite projects in Amazon. 00:06:35.280 |
it has become engaging for our customers as well. 00:06:38.440 |
We're not there in terms of where we want it to be, right? 00:06:46.280 |
Yes, there is some natural attributes of what you said 00:06:49.360 |
in terms of argument and some amount of swearing. 00:06:57.880 |
- It's more than keywords, a little more in terms of, 00:07:04.400 |
these words can be very contextual, as you can see. 00:07:16.680 |
So we have put a lot of guardrails for the conversation 00:07:23.360 |
and not so much of these other issues you attributed, 00:07:30.320 |
- Right, so this is actually a serious opportunity. 00:07:55.320 |
of where the whole industry is moving with AI, 00:07:58.840 |
there's a dearth of talent in, given the demands. 00:08:02.320 |
So you do want universities to have a clear place 00:08:07.320 |
where they can invent and research and not fall behind 00:08:11.360 |
Imagine all grad students left to, to industry like us 00:08:20.320 |
So this is a way that if you're so passionate 00:08:31.920 |
- So what do you think it takes to build a system 00:08:37.080 |
- I think you have to start focusing on aspects 00:08:41.880 |
of reasoning that it is, there are still more lookups 00:08:51.640 |
and responding to those rather than really reasoning 00:08:59.960 |
For instance, if you have, if you're playing, 00:09:13.280 |
that are being mentioned so that the conversation 00:09:16.520 |
is coherent rather than you suddenly just switch 00:09:22.640 |
and you're just relaying that rather than understanding 00:09:26.160 |
Like if you just said, I learned this fun fact 00:09:29.760 |
about Tom Brady rather than really say how he played 00:09:36.760 |
then the conversation is not really that intelligent. 00:09:51.160 |
because a lot of times it's more facts being looked up 00:09:54.880 |
and something that's close enough as an answer 00:09:59.520 |
So that is where the research needs to go more 00:10:05.840 |
And that's why I feel it's a great way to do it 00:10:10.960 |
working to make, help these AI advances happen in this case. 00:10:19.600 |
What is the experience for the user that's helping? 00:10:23.960 |
So just to clarify, this isn't, as far as I understand, 00:10:34.720 |
and it was like, oh, we're checking the weather 00:10:43.000 |
I don't know, how do people, how do customers think of it? 00:10:52.040 |
And let me tell you how you invoke the skill. 00:10:57.600 |
And then the first time you say, Alexa, let's chat, 00:11:13.640 |
And we have a lot of mechanisms where as the, 00:11:21.040 |
then you send a lot of emails to our customers 00:11:24.120 |
and then they know that the team needs a lot of interactions 00:11:33.280 |
who really want to help these university bots 00:11:37.800 |
And some are just having fun with just saying, 00:11:41.400 |
And also some adversarial behavior to see whether, 00:11:52.720 |
if we talk about solving the Alexa challenge, 00:12:04.840 |
'Cause if we think of this as a supervised learning problem, 00:12:09.560 |
but if it does, maybe you can comment on that. 00:12:19.960 |
- I think that's part of the research question here. 00:12:26.560 |
which is have a way for universities to build and test 00:12:33.160 |
Now you're asking in terms of the next phase of questions, 00:12:38.440 |
what does success look like from a optimization function? 00:12:44.520 |
we as researchers are used to having a great corpus 00:12:50.000 |
then sort of tune our algorithms on those, right? 00:13:10.280 |
where just now I started with giving you how you ingress 00:13:14.600 |
and experience this capability as a customer. 00:13:24.960 |
how likely are you to interact with this social bot again? 00:13:31.240 |
and customers can also leave more open-ended feedback. 00:13:42.000 |
that as researchers also, you have to change your mindset 00:13:45.960 |
that this is not a DARPA evaluation or an NSF funded study 00:13:52.400 |
This is where it's real world, you have real data. 00:13:58.960 |
And then the customer, the user can quit the conversation 00:14:09.160 |
- So, and then on a scale of one to five, one to three, 00:14:12.440 |
do they say how likely are you, or is it just a binary? 00:14:18.280 |
That's such a beautifully constructed challenge, okay.