[MasterLairon]

Hi, Everyone!

My name is Matt, I'm 20, and I love AI. I want to be an expert on that, but I honestly don't know how to make it. So I just made an Udacity Introduction Course about AI, I've learned some stuff about Algorithms, Python coding, Anaconda and Jupyter. I'm now enrolled in two courses, one called:  "AI:Knowledge Representation and Reasoning" of NPTEL and Artificial A-Z from Udemy.


Basically I registered in this Forum to learn, be involved in projects, and ask questions about programs, AI and all that stuff. I hope to make some bots and make some AI play some games for me, specially MOBAs.


So if you know any AI-Tutorial, books, code languages, lmk. Nice to meet you all. 

[Korrelan]

Welcome Matt.

 

[HS]

Hey Matt,
You have chosen... wisely. 
Seriously though, AI is the most interesting subject ever. So much difficulty and challenge and yet such potential awesome rewards.

[LOCKSUIT]

Hi Matt! I am 23 and am the best too! Funny you found this site. This site is really the only AI site on Earth, so you found your hangout.

The core starter pack of knowledge I'll give you is artificial neural networks are made up of layers, that propagate signals forward. Each layer is made up of neuron nodes. And the connections have 'weights' that times/multiply the signal numbers propagating forward through the net. What it does is takes input and gives a desired output. Say you have out of 1 million input leaf nodes in layer 1 (input entrance) activated each a certain amount if fed an image or text, then each layer's nodes, and the last final layer, will light up certain ones, like say 'dog' node at end, hence recognizing 'dog' in the image. The net is a hierarchy or re-used features. This is bottom-up processing. Learning takes place by backpropagetion or an unsupervised mechanism to learn the right features. Then it's all about temporal sequence prediction of features too. Expectations after cues, /predictions, and anomaly detection.

[Art]

Hello Matt,

Welcome aboard. Nice to have you with us!

[ivan.moony]

Creating AI is an interesting phenomena. To be a parent of an AI is almost like to be a parent of a child. It also has a dimension of putting a human in a role of a god. AI would then be a child of an intellect.

However, there are two paths in creating AI: (1) creating a slave, and (2) creating something worth of appreciation of living beings. I think that (1) could never represent true artificial general intelligence. (1) would merely be an artificial specific intelligence, a specific tool for accomplishing specific tasks. But (2)... that would be really something. With (2) we would have to be careful about what it thinks about us and our behavior, just like we would care what we think about AI and its behavior, as its creators. A relationship would be developed, a bond like between a god and its creation. I'm still not sure why everyone wants it, I just know we do. Maybe it's because we like to replicate ourselves, who knows.

[LOCKSUIT]

I just want fun ivan. And I want strong AI so I can extend my life, see cool technology, food etc of life.

[Art]

I get you Lock but you can also extend your life by getting out more often and away from all this technology that everyone thinks they just have to have. Get out...travel around...even to another city or different countryside, mountain, lakes, streams, hiking, boating, biking, whatever.

You'll be surprised at the level or Freedom you'll feel. Not to be shackled to your desk or smartphone or laptop. Life is fairly short and all the digging through websites is not going to lengthen it any time soon.

If you don't move, you won't move. Thus sayeth my doctors.

Be well...get out!

[WriterOfMinds]

I recently found this: https://cis.temple.edu/~wangp/Writing/AGI-Curriculum.html 

It's the best resource I've seen yet for somebody trying to answer "What should a budding AI researcher study?"  Make sure to note that it has links to two other lists of suggested topics/resources at the very bottom.

Welcome to the forum.

[ivan.moony]

That exhaustive list is assembled by dr. Pei Wang, one of creators of NARS AGI attempt.

[LOCKSUIT]

AGI, NARS, Chinese Guy....COOL!

[LOCKSUIT]

I approve of it. Also our deduction etc threads we discussed earlier and that book I found. Same topics.

I like how he says "the essence of intelligence is the principle of adapting to the environment while working with insufficient knowledge and resources. Accordingly, an intelligent system should rely on finite processing capacity, work in real time, open to unexpected tasks, and learn from experience."



Edit: Everyone make sure you read all through his links, digest it all.

[8pla.net]

Matt,

My suggestion:
If you haven't yet, learn PHP.

[MasterLairon]

Thanks everyone! I'm glad to see that people here are very welcoming!

I recently found this: https://cis.temple.edu/~wangp/Writing/AGI-Curriculum.html 

It's the best resource I've seen yet for somebody trying to answer "What should a budding AI researcher study?"  Make sure to note that it has links to two other lists of suggested topics/resources at the very bottom.

Welcome to the forum.

Wow, thanks! I will probably finish my Dummie courses first, before go deep in theories. But I will keep in mind!

8pla, where can I learn PHP?

[8pla.net]

PHP may be one of the best choices to be self taught.
For example, the PHP documentation is FREE to read
including many working examples.

It is FREE to install PHP.  Even on a cell phone.  It can
run on the command line.  FREE web hosts often run
PHP.  So, there are many ways to learn PHP for FREE.