AAAI 2015

AAAI 2015

29th AAAI Conference

Name

Twenty-Ninth AAAI Conference

Organizer

AAAI (Association for the Advancement of Artificial Intelligence)

Location

Hyatt Regency - Austin, Texas

Date

January 25th to 30th, 2015

Role

Student Scholar, Volunteer

About

The Twenty-Ninth AAAI Conference on Artificial Intelligence (AAAI-15) was held January 25–30, 2015 at the Hyatt Regency in Austin, Texas, USA.

The purpose of the conference is to promote research in artificial intelligence (AI) and scientific exchange among AI researchers, practitioners, scientists, and engineers in affiliated disciplines.

I was accepted to the conference as a Student Scholar.

The Student Scholar Program provides partial travel support for students who are full-time undergraduate or graduate students at colleges and universities; are members of AAAI; submit papers to the conference program or letters of recommendation from their faculty advisor; and submit scholarship applications to AAAI.

Adventures with AAAI

To note…

This is more of a telling of a personal journey than a technical description or overview of what the conference entails. In other words, it’s meant to be informal so do not take it too seriously (kind of like everything else in blogging).

Day 1 - Sunday, January 25 - ARRIVAL

Much jet lag. So ditzy. I made my seat 37A for both flights to make things easy. No problems anytime along the trip so it’s just passing through the motions.

Using Super Shuttle to go to the hotel, which is cheaper than a standard taxi and also feels safer and less awkward because there are other people driving with you.

Upon reaching the hotel, I go to the concierge to check in. The irony? More people needed the human in order to check in than the kiosks where you could automatically check in yourself. This was an AI conference where people praise automation. A fellow attendee made light of the situation. I giggled.

After checking in and receiving my conference badge along with some AAAI swag, I went to my room and napped. Later, I indulged myself and ordered room service which included a salad, tortilla chips, and a chocolate caramel tart. They delivered it to the wrong room, so I got my food for free. First win.

After many failed attempts to connect to the Internet, I called the front desk and asked them to help me out. They incorrectly redirected me to the Hyatt in San Diego and told me there has been some problems with Macs and connecting in general. However, after the Hyatt in San Diego tried to redirect me to the right place, the phone just kept ringing tunes while on hold. I hung up.

I call later and tell them I’m still having trouble after trying to debug it myself. They apologize and to make up for it, they give me complimentary breakfast for my entire stay. Second win. They let me know that they’ll connect me with the IT manager sometime the next morning.

Time for sleep. Third win.

Day 2 - Monday, January 26

Free breakfast at the Southwest Bistro Note to self: get more pineapple next time

First official AI event was a tutorial called RoboCup, which (surprise, surprise) talked about some of the leagues involved in the RoboCup competition. Stayed for RoboCup Soccer and RoboCup Rescue. Both presentations were enhanced through video footage, but I think I preferred the soccer presentation more just because the speaker went a little faster and it easier to grasp every concept.

Went on to another tutorial called Artificial Intelligence and Technological Unemployment.

I really enjoyed this talk and after it ended there was some debate amongst the audience as to whether or not the so called paradise humans were heading towards where machines essentially do all their work and humans are free to pursue whatever their little hearts desire was a good or bad thing.

The presenter thought this was not something we should be looking forward to and that the idea of losing our chance to work and feel like we’re contributing everyday takes away our humanity while those in the audience felt differently and felt that the chance to pursue what we want wasn’t such a terrible thing at all.

I was actually first introduced to this whole idea of increased technological unemployment through the advancement of AI in the book The Last Firewall by William Hertling. In this little future, the protagonist likes the idea of being in a position to learn whatever she wants and most people have neural implants anyhow which makes these sort of amped up humans with little to worry about very commonplace. The book definitely cast technological unemployment in a positive light.

Next up was another tutorial, this time it was Voting Rules for AI. The first speaker covered all the various methods there are for voting and basically concluded that for 2 candidates, the best voting method would always be plurality (so majority wins) but with more than 2 candidates, there just isn’t any best method at all and boohoo.

I also briefly visited the RoboCup soccer exhibition but there was not a game going on, so I left.

Finally, there was the fancy reception at the Gates-Dell Complex over at UT Austin (so many orange shirts!).

The complex basically houses the entire CS department and is definitely a very modern, clean, and generally tech savvy arena. I explored on my own and later took a tour.

I almost fall asleep on the shuttle back to the hotel.

Day 3 - Tuesday, January 27

More free breakfast and this time with more pineapples. Breakfast seems to taste sweeter when you’re saving $17.

The official welcome and opening remarks are today. The presentation got a little staticky.

About 15 minutes after the opening remarks, Oren Etzioni, an invited speaker and also the CEO of Ai2, goes ahead and delivers his talk. What was really cool was when he showed his sneak-peek demo of Semantic Scholar which is this search engine that allows users to find key survey papers about a topic or to produce a list of important citations or results in a given paper. The user interface was a lot more sophisticated than in the other projects I’ve looked at so far and was what really made it stand out to me at least. He also talked about Jeff Hammerbacher’s quote: “The best minds of my generation are thinking about how to make people click ads. That sucks.” and how he wanted the community to leverage AI for the better good. More discussion on that quote here. He also discouraged focusing on one particular method in AI for research/projects and instead encouraged that people try and build integrated systems that use all the different methods instead, even it does take longer and is generally more difficult to perform.

After that, there was a short break and then technical sessions started up so I went to the Machine Learning focused session and that lasted quite a while. To be fair, most of the material was a bit over my head being a newbie to all this but to be fair, the presenters weren’t aiming to appeal to a broader audience.

For the lunch break, I visit the MakerSquare location in Austin, which is conveniently located about 20 minutes away from the hotel walking. I get a quick tour of the space and am able to ask Amanda, the admissions director, a few questions about future changes in MakerSquare since they’ve recently been acquired by Hack Reactor. I’m considering whether I should attend MakerSquare or Fullstack Academy for the summer or if I should just forget about bootcamps altogether and focus on an internship instead. It’ll be a very difficult decision, regardless.

I attend the Shakey celebration for a short while and was amused and excited by how far we’ve gone since this guy.

Poster and Demos session is the last event for the next 3 nights. It’s where researchers showcase their work or demo their project. It’s also where various companies/institutions will gather to showcase what they’re currently working on as well. I walk around, eat some food, and get to bed.

Day 4 - Wednesday, January 27

Wednesday was probably my favorite day of the conference. This was probably because I was acting as a volunteer throughout the evening.

Early in the morning were the What’s Hot Talks which basically introduced innovative breakthroughs in a certain field within A.I. This morning, it was Human-Computer Interaction (or HCI). It went a little slowly for being only 15 minutes long, but afterwards I got to see my first Ph.D student presentations in the Robotics Student Fellowship Talks session.

I think perhaps at this point I was growing tired of seeing older men present and found younger people showcasing their research more appealing.

There were about 6 presentations in total, 3 of which weren’t that much over my head where I could really go ahead and comprehend the majority of what they were speaking about.

Next up, after an ML session, an invited talk from Geoffrey Hinton, and lunch, there was the community meeting. I was to act as a volunteer and pass out little tickets that would act as a complimentary drink for anyone who was going to attend the meeting. It was meant to serve as an incentive, but apparently the marketing wasn’t very well done. Regardless, most people appreciated the gesture.

I listened more towards the end and the president of triple-AI mentioned that the conference was trying to get more minorities and women in general to attend and that they had already implemented a system of child-care at the conference where it was possible for attendees who were also mothers to have their children around (which is why I saw one or two little people, no doubt).

To be fair, the conference was about as diverse as the tech community: mostly whites, Asians, and some South Asians. Mostly men too. On average, it seemed that in a room of about 20 people, there were maybe 5 women. However, in contrast to most of the industrial scene (at least with start-ups), the majority of the men were older and not younger.

Also, since I’m already on this topic, I will say that I saw less than 4 people that would pass as black. The same idea applies to Hispanics like me, but then again, it’s always harder to tell who’s Hispanic just because they can look like anybody else.

Having said that, I will commend them on having a lot of international diversity. I saw people coming from Israel, Australia, India, China, and Germany from my limited experience.

Anyways…

Following the community meeting, there was the Poster and Demos session where I followed the same routine as yesterday: walk around and eat the food.

And for the final event of the evening where I would again be volunteering: Games Night. Here I also had to help pass out tickets. Each attendee (unless they didn’t want to participate at all) would receive one, either to enter in to play the Price is Right (AAAI Version) or be entered in to receive prizes in the raffle at the end. The moderator was pretty snazzy so it was fun either way.

A lot of people didn’t know what the Price is Right was (frankly, neither did I) and I ended up spilling the same story every time: the Price is Right is where you estimate the price of certain things and it’s all about estimation and probability. There was a larger turn-up than I expected that there wasn’t any space left at the climax. There were mini mathematically oriented games beforehand and during the primary Price is Right game. There were also mini iPad games. The winner ended up with a backpack full of swag from the conference. Then there was the raffle. One of the organizers of the event even let me have a free IBM Watson t-shirt which looks a little something like this:

I took pictures of downtown too just because it was such a beautiful scene outside.

I have to wake up early in the morning, so once the event officially ends at 10 pm, I ask if I can leave to go to bed.

Day 5 - Thursday, January 28

Early in the morning, there is a speaker from the IBM Watson Research Center called Meinolf Sellmann.

I recall that he’s one of the fellows I traveled with through the Super Shuttle service on my way to the conference. I remember because some lady asked if his accent was Scandinavian (it was actually German).

The talk is all right and he goes over his primary accomplishment with the SAT Solver.

There’s another What’s Hot Talk. This time, it’s a lot more interesting than HCI simply because someone talks about artificially intelligent Angry Birds. Who knew?

//www.youtube.com/watch?v=AplG6KnOr2Q

Then there was a short period for the awards for the video competition. Needless to say, it was pretty awesome seeing an artificially intelligent Mario!

After a session on NLP, it’s time for more student fellowship talks.

Ph.D students never fail to be interesting. Well, most of the time.

It was fun to learn more about affordance models, because it reminded me of Baxter, who I’m familiar with, even though the student was using Curi from Georgia Tech.

Someone from CSAIL at MIT also talked about human-robot collaboration and concluded that humans don’t think any less of their robot counterparts assuming they’re highly autonomous. Go figure.

There was also someone who talked about multi-agent rendezvous. This talk was a little harder to understand, but her research certainly looked promising.

During lunch break, while I’m in the elevator heading to my room, some couple asks a butler what’s going on. I let them know that the triple AI conference is going on. “What does that stand for?” “The Association for the Advancement of Artificial Intelligence"( It’s always such a mouthful). They go on to talk about how all the ‘computer nerds’ and ‘laptop people’ must be here (no joke). They leave the elevator. I chuckle quite a bit, but then feel a little sad.

Then there’s the debate, something I was looking forward to. It was a debate on autonomous weapons. I figured it would be intense, but to quote someone I overheard, it was “not satisfying”.

One of the two debaters had a PowerPoint to accompany his arguments and there was so much text on each slide with him speaking relatively quickly, it was hard to grasp what his main idea was. Basically though he was arguing that there was such a thing as implementing an ethically sound robot and that while humans would never be out of the question, it was possible to have autonomous robots/weapons that could act more ethically moral than their human counterparts who could potentially carry emotional baggage and can act irrationally under dangerous environments that according to the speaker, “humans weren’t made to be in”. I agree with this argument, but I also recognize that I’m a pretty idealistic person.

Unfortunately, I noticed most of the audience looking at their laps.

Then there was the other debater who was arguing that R&D towards autonomous weapons should be banned completely. He had no PowerPoint to accompany his arguments and was staring at his paper practically the entire time. Maybe I was expecting better debate practices from both speakers but at least it was clear they both knew a lot about their particular stances.

After the rebuttals and one or two questions from the audience, I leave to attend the last Poster and Demo session where, like with the last 2 days, I follow the same routine of walking around and eating food.

I’m particularly pleased with the way this session turns out because I was able to strike a conversation with somebody working with crowd-sourcing data in Bitcoin. It was random, but I came away knowing more about the mechanizations behind Bitcoin and why it’s obvious there is a lot of risk involved.

Also, there was wild mushroom pasta, which was quite possibly the best thing I ate since I arrived.

Day 6 - Friday, January 30 - DEPARTURE

Having reserved the Super Shuttle service the night before, I wake up early so I can check out and catch the shuttle at 8 am.

The bill for the hotel comes out to $1086.75Yikes. As I’ve mentioned before, I’m just glad that I have that VSGC scholarship and a perfect eBay seller’s reputation…

The man running the shuttle, who I find out later is Columbian because the other fellow he picked up from the Hyatt is also Columbian, doesn’t have a credit card scanner/device so I end up with a complimentary ride to the airport. Win.

We pick up a lady from another hotel, who is also Hispanic, and for the first time in a while, I’m surrounded by 3 Latinos. It’s about time. I had been expecting more Hispanics being in Austin and all.

When I pull out my phone to check in, there are again some surprised faces. I assumed incorrectly that a lot more people would have their phones ready to check in at the Austin airport, having accustomed themselves to the growing functionality of their precious phones. Alas, there were only slightly more.

Slightly Irrelevant : When I’m waiting to leave the plane at arrival I notice some patterns with my flights and I notice how neat it would be if autonomous vehicles were driving the luggage around instead and mini-robot arms were packing them in. In both flights, I’m in front of a mother and child and in both, there is a gay, white male flight attendant (well, I presumed they were). And in both, of course, I sat in 37A.

I finally arrive back at home sweet home, my little home by the sea, my Norfolk. My mom and sister pick me up, I feed Nefer some treats the minute she greets me, and my mom has some mushroom pasta fixed for me.

All in all, it’s been rather delightful.

If Prezi does not work, check it out here!

Award

January 2015 - Student Scholar Program Awardee

AAAI is pleased to announce the continuation of its Student Scholarship Program for 2016, which is sponsored by the AI Journal. The Student Scholar Program provides partial travel support for students who are full-time undergraduate or graduate students at colleges and universities; are members of AAAI; submit papers to the conference program or letters of recommendation from their faculty advisor; and submit scholarship applications to AAAI by November 30, 2015. In addition, repeat scholarship applicants must have fulfilled the volunteer and reporting requirements for previous awards. In the event that scholarship applications exceed available funds, preference will be given to students who have an accepted technical paper, and then to students who are actively participating in the conference in some way. However, all eligible students are encouraged to apply. After the conference, an expense report will be required to account for the funds awarded.

The Student Volunteer Program is an essential part of the conference and student participation is a valuable contribution. Volunteers will support AAAI organizers in Phoenix, Arizona. In 2016, a limited number of complimentary technical program registrations will be available for students who volunteer during the conference. Preference will be given to participating students for the volunteer positions. Local students or students not requiring travel assistance can apply for the Volunteer Program if openings are available. AAAI membership is required for eligibility. The deadline for volunteer applications is November 30, 2015.

Subscribe to my newsletter Today

In my newsletter, I share personal and career learnings, advice, collaboration opportunities, and more.