The Master Algorithm: How the Quest for the Ultimate Learning Machine Will Remake Our World

by Pedro Domingos

Homo sapiens is the species that adapts the world to itself instead of adapting itself to the world.

Some things, try as we might, are just unpredictable. For the vast middle ground between the two, there’s machine learning.

But finding correlations is to machine learning no more than bricks are to houses, and people don’t live in bricks.

(As Richard Feynman said, “What I cannot create, I do not understand.”)

Scientists make theories, and engineers make devices. Computer scientists make algorithms, which are both theories and devices.

Inevitably, however, there is a serpent in this Eden. It’s called the complexity monster.

Learning algorithms are the seeds, data is the soil, and the learned programs are the grown plants.

Machine learning is sometimes confused with artificial intelligence (or AI for short). Technically, machine learning is a subfield of AI,

Should the link at the bottom of a page be red or blue? Try them both and see which one gets the most clicks. Better still, keep the learners running and continuously adjust all aspects of the website.

machine learning will bring about not just a new era of civilization, but a new stage in the evolution of life on Earth.

All knowledge—past, present, and future—can be derived from data by a single, universal learning algorithm.

Thus it seems that evolution kept the cerebellum around not because it does something the cortex can’t, but just because it’s more efficient.

if everything we experience is the product of a few simple laws, then it makes sense that a single algorithm can induce all that can be induced.

the Master Algorithm has at least as many skeptics as it has proponents. Doubt is in order when something looks like a silver bullet.

Without machine learning, the number of ideas needed to build an intelligent agent is infinite. If a robot had all the same capabilities as a human except learning, the human would soon leave it in the dust.

There’s a saying in industry: “Listen to your customers, not to the HiPPO,” HiPPO being short for “highest paid person’s opinion.”

As Isaiah Berlin memorably noted, some thinkers are foxes—they know many small things—and some are hedgehogs—they know one big thing.

Teach the learners, and they will serve you; but first you need to understand them.

The future belongs to those who understand at a very deep level how to combine their unique expertise with what algorithms do best.

Their master algorithm is inverse deduction, which figures out what knowledge is missing in order to make a deduction go through, and then makes it as general as possible.

The connectionists’ master algorithm is backpropagation, which compares a system’s output with the desired one and then successively changes the connections in layer after layer of neurons so as to bring the output closer to what it should be.

The evolutionaries’ master algorithm is genetic programming, which mates and evolves computer programs in the same way that nature mates and evolves organisms.

The solution is probabilistic inference, and the master algorithm is Bayes’ theorem and its derivates.

The analogizers’ master algorithm is the support vector machine, which figures out which experiences to remember and how to combine them to make new predictions.

Machine learners, like all scientists, resemble the blind men and the elephant: one feels the trunk and thinks it’s a snake, another leans against the leg and thinks it’s a tree, yet another touches the tusk and thinks it’s a bull.

How can we ever be justified in generalizing from what we’ve seen to what we haven’t?

How about we just assume that the future will be like the past? This is certainly a risky assumption.

This is the machine-learning problem: generalizing to cases that we haven’t seen before.

In machine learning, examples of a concept are called positive examples, and counterexamples are called negative examples.

Learning is forgetting the details as much as it is remembering the important parts.

Overfitting is the central problem in machine learning.

Rulea that perfectly apply to the given data. Like the spam filter that marks as spam only if is an exact copy of a previously flagged email.

Thus a good learner is forever walking the narrow path between blindness and hallucination.

David Marr argued that every information processing system should be studied at three distinct levels: the fundamental properties of the problem it’s solving; the algorithms and representations used to solve it; and how they are physically implemented.

knowledge-based systems had trouble dealing with uncertainty,

Another difference between symbolist and connectionist learning is that the former is sequential, while the latter is parallel.

In particular, the computer can use speed to make up for lack of connectivity, using the same wire a thousand times over to simulate a thousand wires.

evolution as a whole has no known purpose.

??? Maybe that is why there are so many branches, there is not one "best" animal. I believe competition is key.

Sex just seems to be the end, rather than the means, of technological evolution.

Of all the possible genomes, very few correspond to viable organisms. The typical fitness landscape thus consists of vast flatlands with occasional sharp peaks, making evolution very hard.

“Nature” for a computer is the program it runs, and “nurture” is the data it gets.

I like this analogy for computers

Bayesians’ answer is that a probability is not a frequency but a subjective degree of belief.

machine learning is the art of making false assumptions and getting away with it.

lol

Kalman filter.

As one wag put it, space is the reason everything doesn’t happen to you.

According to Ray Kurzweil, the Singularity begins when we can no longer understand what computers do.

the two main subproblems in analogical reasoning: figuring out how similar two things are and deciding what else to infer from their similarities.

If Cope is right, creativity—the ultimate unfathomable—boils down to analogy and recombination.

Above all, even though children certainly get plenty of help from their parents, they learn mostly on their own, without supervision, and that’s what seems most miraculous.

Suppose the entities in Robby’s world fall into five clusters (people, furniture, toys, food, and animals), but we don’t know which things belong to which clusters. This is the type of problem that Robby faces when we switch him on.

There is where feedback helps. If something hurts me is bad, if it gives me pleasure is good. These two feedback experiences are not supported by algorithms.

Humans do have one constant guide: their emotions. We seek pleasure and avoid pain.