Artificial General Intelligence is Nowhere Close To Being a Reality (venturebeat.com)

Three decades ago, David Rumelhart, Geoffrey Hinton, and Ronald Williams wrote about a foundational weight-calculating technique -- backpropagation -- in a monumental paper titled "Learning Representations by Back-propagating Errors." Backpropagation, aided by increasingly cheaper, more robust computer hardware, has enabled monumental leaps in computer vision, natural language processing, machine translation, drug design, and material inspection, where some deep neural networks (DNNs) have produced results superior to human experts. Looking at the advances we have made to date, can DNNs be the harbinger of superintelligent robots? From a report: Demis Hassabis doesn't believe so -- and he would know. He's the cofounder of DeepMind, a London-based machine learning startup founded with the mission of applying insights from neuroscience and computer science toward the creation of artificial general intelligence (AGI) -- in other words, systems that could successfully perform any intellectual task that a human can. "There's still much further to go," he told VentureBeat at the NeurIPS 2018 conference in Montreal in early December. "Games or board games are quite easy in some ways because the transition model between states is very well-specified and easy to learn. Real-world 3D environments and the real world itself is much more tricky to figure out ... but it's important if you want to do planning."

Most AI systems today also don't scale very well. AlphaZero, AlphaGo, and OpenAI Five leverage a type of programming known as reinforcement learning, in which an AI-controlled software agent learns to take actions in an environment -- a board game, for example, or a MOBA -- to maximize a reward. It's helpful to imagine a system of Skinner boxes, said Hinton in an interview with VentureBeat. Skinner boxes -- which derive their name from pioneering Harvard psychologist B. F. Skinner -- make use of operant conditioning to train subject animals to perform actions, such as pressing a lever, in response to stimuli, like a light or sound. When the subject performs a behavior correctly, they receive some form of reward, often in the form of food or water. The problem with reinforcement learning methods in AI research is that the reward signals tend to be "wimpy," Hinton said. In some environments, agents become stuck looking for patterns in random data -- the so-called "noisy TV problem."

Comments

Break It Down

[JBMcB]

Put simply - most of the "Artificial Intelligence" you hear about in the news is really fancy pattern matching. So you can have software that can recognize voice commands, or faces in pictures, or general patterns in data.

What you don't have, and aren't even close to, are computers that can "think." That is, put different sets of data together in arbitrary ways and make sense of it. You can't feed in a bunch of musical information to a computer and have it spontaneously generate music. You can't feed in a bunch of economic data and have it decide that certain regulations are required to achieve some economic goal - unless someone specifically programs it to do so.

The underlying reason is computers lack any way of attaining "common sense." If you tell a computer a person is in a room, the computer has no concept of what you are talking about but will dutifully note that a person is in a room. To a computer that could mean the person is occupying all the space in the room, that the person is in every room that exists, that the person is in the room AND outside the room, or that a person IS a room. In actuality, the computer makes no inference beyond "something called a person is in something called a room, whatever that means."

Re:Break It Down

[ShanghaiBill]

Put simply - most of the "Artificial Intelligence" you hear about in the news is really fancy pattern matching.

Put simply - most of the "Human Intelligence" you see is really fancy pattern matching as well.

Re:Break It Down

[hazardPPP]

Put simply - most of the "Artificial Intelligence" you hear about in the news is really fancy pattern matching. So you can have software that can recognize voice commands, or faces in pictures, or general patterns in data.

What you don't have, and aren't even close to, are computers that can "think." That is, put different sets of data together in arbitrary ways and make sense of it. You can't feed in a bunch of musical information to a computer and have it spontaneously generate music. You can't feed in a bunch of economic data and have it decide that certain regulations are required to achieve some economic goal - unless someone specifically programs it to do so.

The underlying reason is computers lack any way of attaining "common sense." If you tell a computer a person is in a room, the computer has no concept of what you are talking about but will dutifully note that a person is in a room. To a computer that could mean the person is occupying all the space in the room, that the person is in every room that exists, that the person is in the room AND outside the room, or that a person IS a room. In actuality, the computer makes no inference beyond "something called a person is in something called a room, whatever that means."

Wasn't this obvious to anyone who has studied neural networks and deep learning? I mean I would shake my head each time someone would claim that deep learning would create functioning computer "minds".

Yes, it's obvious that our brains do a lot of very efficient pattern recognition (that often misfires, but when it does, it usually errs on the side of caution - clear evolutionary adaptation). However how can anyone in their right mind be so reductionist as to think that ALL that are brains do is fancy pattern recognition?

It's similar to AI hype of previous ages, when people thought that logical programming languages would create AI, as if human intelligence was logic only. The use of logic is only a subset of human intelligence and we use it less often than we like to think. Formal logic is a human construct, and replicating human-type thinking using formal logic only was never going to work. With deep learning we went completely the other way, throw enough artificial neurons and data at it and magically a mind will emerge. All this time we don't truly understand what a "mind" is in its totality, which makes replicating it in computers - things built to very deliberately follow precise instructions - like, really hard.

Computers beating humans at chess or go does not mean AI has arrived. Chess and go are human inventions, they are games invented with very clear and defined rules. Therefore it is possible to create other human constructs (computer programs) that can exploit these rules and large amounts of computational power to beat humans. It can be very hard, the solution can be very impressive, but it does not mean we have AI. In fact there is no rule about transferring chess skills into other, unrelated domains (Fischer and Gasparov come to mind, both not being quite sane in their post-chess careers), and the same goes for other very specific skills. Training a computer to be very good at face recognition says nothing about "AI", really.

Humans suffer from explanatory reductionism based on the dominant technological paradigm of the time. We try to explain the entire world using things which we know well. When we were an agricultural society, the world was a flat disk held up by giant pack animals. When Newton's theories revolutionized science and the industrial revolution revolutionized the economy, we saw the universe as a clockwork mechanism. After the computer revolution, we think everything can be reduced to some form of computation (and some posit that we are in fact living in a computer simulation).

Sense

[JBMcB]

Put simply - most of the "Human Intelligence" you see is really fancy pattern matching as well.

That's a big part of it, but there's some "secret sauce" that lets organic brains combine patterns in new and different ways that AI researchers haven't been able to crack. Whatever it is, it's more than just matching patterns.

About goddam time ...

[CaptainDork]

... because a shit load of us have been yakking about this for years.

"Artificial intelligence," will be a reality when your smart device says, "Sorry. I'm just not in the mood right now."

Re:Algos != intelligence, artificial or otherwise.

[ShanghaiBill]

Until we have a proper definition for intelligence my pet rock qualifies.

Here is the proper definition of intelligence:

Intelligence: The ability to formulate an effective initial response to a novel situation.

Each word is important:
1. Intelligence is an "ability" not a mechanism. An entity that behaves intelligently is intelligent. internal mechanism is irrelevant.
2. Intelligence is the ability to "formulate" a plan, not to physically act on it.
3. A response is effective if meets an objective criteria.
4. It is the "initial" response that counts. Success achieved by a long term random process, including evolution over multiple generations, is not intelligence.
5. It is the response to "novel" situations that is the measure of intelligence. It is not just rote application of a solution that worked in the past. Memory and learning are important components of intelligence, but an intelligent entity can see how a past solution may or may not apply, and how to modify it for the new situation.

Your pet rock doesn't qualify.