HN Theater @HNTheaterMonth

I think most people would probably not like your approach, and think of it as way too sci-fi and idealistic given the limitations of current tools. They would suggest studying AI a lot more before you attempt to really think of anything. Things that seem easy can be hard, things that seem hard can be easy. And it's not clear that we will ever really recognize superintelligences...if we program something, then we know how it works, and if we know how it works, we'll just say "Oh, it's an algorithm" and not ever appreciate the extent of machine capabilities.

I do like how you handled "consciousness" though by redefining it to mean 'examine your memory and then learn from it'; artificial intelligence is really just one big word game and if you can change your definitions properly to something achievable, then you might actually achieve them. Then again, you actually have to do it and that seems more difficult than talking about it. Most people will likely disagree with your "consciousness" definition and will prefer a more vaguer version that would be impossible to be built...

My suggestion is two-fold:

1) Learn AI first and find out its tools and limitations. Struggle through those boring AI lectures, or see if you can find any GitHub Machine Learning libraries so that you can start working. Also, try some side projects where you use AI to solve real-world practical problems, before you attempt to figure out how to induce the existence of an AGI. I'd even suggest dropping the existence of an AGI (there's no guarantee it can even exist anyway) and focus on a specific part of AGI: inducing emotion or consciousness by itself might be "pretty cool", and if you can do that without needing an AGI, and be able to apply them to real-world problems, then people might even accept those algorithms as being valid (though they may avoid the pseudo-scientific terms we're applying to them right now).

2) Look at the "Mario Lives" videos[1], where programmers were able to successfully represent emotional states and conversations in an AI that plays Mario. May not match your ideas about emotion and learning, but it is a sign that you can do anything you want so long as you change the definitions of words to be what's practical.

[1]Video 1: Mario Having Emotional States - https://www.youtube.com/watch?v=AplG6KnOr2Q

Video 2: Mario Becoming Social As Well - https://www.youtube.com/watch?v=ltPj3RlN4Nw

⬐ sharkhacks

Thanks for your feedback Tariq. To give you some context:

-I have taken AI as a course in the past (college 10 years ago), and I'm a computer engineer so I have the basics of ML,AI,etc. and I'm familiar with the libraries that are out there and the body of knowledge.

- I find that when you go deep into a particular topic or field and read lots of research you tend to become immersed and start to think like everyone else in the field. There is some wisdom to Zen's mind beginner's mind and I wanted to let that Beginner's mind explore freely and express its thoughts before my reimmersion in the research.

- I will definitely look at Mario Lives video thanks for pointing it out.

- I guess I was looking for someone to challenge the ideas as ideas more than challenge my credentials / approach but it's all fair.

- I'm not interested in solving a particular problem at this point in time in AI. I think everyone is taking this angle and everyone is trying to do this. I'm already in Tech, in a fairly good position, and solving real customer problems. I'm simply following my curiosity and a more grandeur goal and doing it just for the sake of following my curiosity.

⬐ ilyaeck

Strictly speaking, this is hardly AI, but a rather traditional approach of handwritten grammars.

⬐ Houshalter

It involves a lot of techniques which are typically called AI. E.g. planning and learning simple models of the world. The best word is "weak AI" or "narrow AI".

⬐ rcfox

"AI is whatever hasn't been done yet." http://en.wikipedia.org/wiki/AI_effect

⬐ jacquesm

In this case it has been done quite a few times (and better).

⬐ dmazin

Would love to see some examples!

⬐ jacquesm

This is absolutely an impressive demo, but the way to look at it is as a 'mash-up' of a variety of components that have all been demoed before in isolation and usually with a lot more attention to the technology behind it but a much less slick presentation.

So kudos to the people that put this together but it is mostly the mashup that is new, not the individual pieces and those pieces have demos all their own that make for very interesting reading and viewing (where applicable).

I should do a blog post on this, but that will be a ton of work to properly represent the state of the art (which this demo does not).

⬐ maratd

http://alice.pandorabots.com/

⬐ listic

    judge: The cat is green, the dog is blue. 
    ALICE: Is green the dog is blue also cat? 
    judge: What color is the dog? 
    ALICE: orange.

I am not impressed.

⬐ maratd

Nor should you be. But I think Alice will do better than Mario. All Mario does is talk about his feelings. Alice can do that too.

⬐ modeless

The speech recognition and synthesis are primitive compared to any smartphone assistant these days. The natural language processing is about equivalent to SHRDLU from the 1960s [1]. It turns out that this approach based on manually constructing syntax trees and applying simple logic can make some fun demos but is ultimately a dead end in terms of making systems that are actually useful, as was discovered in the "AI winter".

The part that controls Mario looks similar (if not identical) to this: http://aigamedev.com/open/interviews/mario-ai/

If you want to see the state of the art in using AI to play video games, look no further than "Playing Atari with Deep Reinforcement Learning" [2], where a single general AI system learns to play many different games. The generality is what makes it impressive. Its only inputs are pixels and score, and its only outputs are joystick and button state, just like a human player. This makes it unlike these Mario systems which are hand programmed very specifically for Mario, and use special instrumentation of the game state that skips pixels entirely.

[1] http://en.wikipedia.org/wiki/SHRDLU

[2] http://arxiv.org/abs/1312.5602

⬐ nl

Note that the Atari playing system is what got Google interested in buying Deep Mind (the company behind it). It was a pretty significant advance on the state of the art at the time.

Deep Mind sold to Google for around $500M

⬐ mietek

So, SHRDLU and what else?

You do know that SHRDLU is still unique enough to warrant multiple resurrection attempts?

http://www.semaphorecorp.com/misc/shrdlu.html

⬐ SunShiranui

I've always been interested in the use of natural language recognition in games. Does anyone have some good resources about the topic?

⬐ javert

This is a shrimp treadmill if I ever saw one. [1]

[1] http://chronicle.com/blogs/conversation/2014/11/13/how-a-47-...

⬐ Animats

There have been games with NPCs smarter than that. There have been text adventures with sentence understanding at that level. It's a cute demo, but calling it a "living and conversing agent" is way too much.

"Mario, exit the game." "Do you really want to exit the game?" So not impressed.

On the speech front, there is now an automated telemarketing agent which is about as smart as most script-driven human telemarketers. That's a somewhat scary development.

⬐ larrydag

I think most of these comments are missing the point. As I understood the video they are trying to create an AI that plays the game. The AI will respond to its experiences as it learns about the game. The initial Mario state is that it knows nothing then as it learns its environment it will adapt to figure out how to play.

This seems like a pretty bold project. I'm curious if they can get the AI to win the game.

⬐ wodenokoto

> "Mario, exit the game." "Do you really want to exit the game?" So not impressed.

This depends on how the response is generated. If it's a script, then of course there is nothing impressive about it, but if parts of the system can recognize importance in action and generate a sentence that asks for confirmation for important actions then it is a completely different ballgame.

> On the speech front, there is now an automated telemarketing agent which is about as smart as most script-driven human telemarketers.

Samantha west is human operated. Some dude is just playing back prerecordings that hopefully fit the conversation. It has absolutely nothing to do with AI.

http://newsfeed.time.com/2013/12/17/robot-telemarketer-saman...

⬐ thewarrior

Could give a few examples ?

⬐ new299

The telemarketing robots that are scripted well enough to continually deny that they're robots are somehow the scariest/most impressive for me:

https://www.youtube.com/watch?v=unRDMy7O62o

⬐ thewarrior

I think that's a person playing back pre recorded sounds. AI isn't advanced enough to impersonate a telemarketer.

⬐ woodman

It is. Offshore telemarketing - no more fake localized names and accents, just hire a person who understands spoken english and install the MidwestFemale 3.68 module.

⬐ chadscira

I had no idea they were using this type of tech already, the second call is so smooth.

⬐ SG-

sounds like Lenny which is an anti telemarketer SIP bot you can forward calls to:

https://www.youtube.com/watch?v=p5IKbFATKLo

⬐ new299

This is amazing. I wonder if anyone has done something similar for email spam.

⬐ SG-

if you're interested in more of these calls (some are quite funny) there's an entire reddit dedicated to it:

http://www.reddit.com/r/itslenny/

⬐ amelius

I usually just say "please hang on, while I get my credit card", and then just leave the phone off the hook...