Monday, February 29, 2016
I deduce from this that computer scientists think that quantum computing is a hot new area of computer science, with huge advances likely in the next 20 years.
I wonder how long they will be able to continue hyping this subject, without getting results.
Friday, February 26, 2016
Quantum Computing as a ServiceYes, the lecture is as bad as it sounds.
Matt Johnson, CEO and Co-Founder, QC Ware
About the talk:
QC Ware will present a business-oriented overview of quantum computing, from the perspective of a start-up that is developing software and tools to address real-world QC use cases. The presentation will cover the strategic and practical considerations of launching a QC software company.
Their business plan is to hope that someone invents a quantum computer, and then to write software programs for it. So far, they have nothing.
Their business is premised on the idea that quantum computers will be commercially successful, and that a lot of highly skilled custom programming will be needed to make them useful.
The plan is hopeless.
I am betting that no one will make a useful quantum computer. It is possible that someone will make an analog (non-quantum) computer that has some commercial utility, and that this company will find a small niche exploiting it for particular applications. But I doubt it.
We must be in a bubble if companies like this can get funding.
The previous week's colloquium was more about how quantum computers work, but it was mainly just a lesson in quantum mechanics.
Wednesday, February 24, 2016
The use of probability theory is widespread in our daily life (gambling, investments, etc.) as well as in scientific theories (genetics, statistical thermodynamics). In virtually all cases, calculations can be carried out within the framework of classical probability theory. A special exception is given by quantum mechanics (the physical theory that describes matter on the atomic scale), which gives rise to a new probability theory: quantum probability theory.Many professors argue that the essence of quantum mechanics is negative probability, or imaginary probability, or interfering probabilities, or something like that.
On the other hand, experts in probability and statistics have found quantum probability to be useless for modeling anything but atomic particles.
However, you only need regular probability theory for quantum mechanics. This has been explained by Jaynes and the Bayesians. Probability is not a physical thing, and is not essential to the theory. It is needed for testing, but that is true of all scientific theories.
Wave can interfere. Probabilities do not.
Monday, February 22, 2016
Quantum weirdness may hide an orderly reality after allBohm proposed an interpretation of quantum mechanic. That means that no experiment can prove it any more correct than the Copenhagen interpretation or any other interpretation.
Often brushed aside like a forgotten stepchild, a 64-year-old theory of quantum mechanics may now share the stage with its more well-regarded siblings. If it holds up, it might lend support to ideas that the universe is improbably interconnected across vast distances.
The theory, by physicist David Bohm, has been resurrected after researchers carried out experiments on photons that seemed to support it.
The behaviour of the quantum world has befuddled physicists for nearly a century. “We have had geniuses working on it and we still have a problem,” says Basil Hiley, a quantum physicist at Birkbeck College at the University of London, who worked with Bohm until the latter’s death in 1992.
Unlike the classical world, with its clockwork precision and pleasing predictability, the quantum world is rife with randomness.
The famous illustration is the double-slit experiment: if you fire photons at two slits, our classical intuition expects each to pass through one or the other slit and hit a screen on the other side, making a single mark indicative of its particle nature. But when you try it, the photons create an interference pattern of light and dark bands on the screen, as if each photon behaved like a wave and passed through both slits simultaneously.
The dominant explanation of such behaviour is called the Copenhagen interpretation, which states that the question of whether a photon is a wave or a particle has no meaning until you make a measurement – and then it becomes one or the other depending on which property you measure. The other favoured explanation is the many-worlds interpretation, under which each possible state of the photon becomes manifest in an alternate world.
But in 1952, Bohm suggested that the quantum world only appears weird because we don’t know enough about its underlying reality. Beneath the quantum weirdness, he said, reality is orderly.
“It’s a very deterministic description, where all the particles in nature have definite positions and follow definite trajectories,” says Aephraim Steinberg of the University of Toronto in Canada.
Many recent experiments have suggested that no such hidden reality exists. However, they have only ruled out a specific class of theories in which the hidden reality of any particle is local, and not influenced by something far away.
Bohm’s ideas involve non-local hidden reality, in which everything depends on everything. In his universe, something happening in a distant galaxy is influencing you right now and vice versa, however minor the effect.
Take the debate over whether an electron is a wave or a particle. Bohm’s theory says that it’s both: an electron is a particle with a definite trajectory, but this path is governed by a wave upon which the electron rides. The wave can also be influenced by other particles, which in turn changes the trajectory of the electron.
If you want to believe that you are being influenced imperceptibly by something in another Galaxy, then Bohm gives you permission to do that, but no experiment will ever back up that view.
The article goes on to describe some variant of the double-slit experiment anyway, and claim that it is compatible with Bohm's interpretation.
“I’m happy to see this resolution. It restores my taste for Bohmian mechanics,” says Steinberg. “We want to bring it back to its rightful place among all other interpretations.”No, only crackpots believe in Bohmian mechanics or many-worlds.
Hiley is impressed by the experiment. “It is a new way of looking at quantum non-locality, which vindicates the Bohm position,” he says.
Sheldon Goldstein, an expert on the foundations of quantum mechanics at Rutgers University in New Jersey, points out that the experiment’s observation of particle trajectories predicted by Bohmian mechanics does not prove that Bohm’s theory on the nature of reality was correct. Such paths can also be explained using other theories, he says.
But Goldstein says there are changes afoot. “After decades and decades, people are taking Bohmian mechanics a little bit more seriously,” he says. “There was a time when you couldn’t even talk about it because it was heretical. It probably still is the kiss of death for a physics career to be actually working on Bohm, but maybe that’s changing.”
These interpretations have some value in showing what interpretations are possible. But that's all. They do not give any better understanding of the world.
If anyone finds evidence of parallel universes or instaneous influences from other galaxies, then we will have to confront that. But lacking such evidence, it is crazy to believe in such outlandish ideas. Believing in ghosts makes much more sense.
The Bohmians would have you believe that quantum mechanics is inexplicably "weird", and is hiding an orderly reality. This is backwards. A world where things in distant galaxies immediatly influence us is not an orderly reality.
The recently discovered gravitational waves took millions of years to get to us. The effect was extremely slight, but getting here immediately would be contrary to ordinary scientific thinking about cause and effect.
The Bohmians want you to throw away such causality. That would be far far weirder than quantum mechanics.
Thursday, February 18, 2016
Lorentz, in the work quoted, found it necessary to complete his hypothesis by assuming that all forces, whatever their origin, are affected by translation in the same way as electromagnetic forces and, consequently, the effect produced on their components by the Lorentz transformation is still defined by equations (4).Newton proved that an inverse square law of gravity gives elliptical orbits, assumed instantaneous transmission of the force. Laplace proved that if the propagation speed were at the speed of light or slower, then Newton's proof would be wrong, and planetary orbits would not be like what we observe.
It was important to examine this hypothesis more closely and in particular to examine what changes it would require us to make on the law of gravitation. That is what I sought to determine; I was first led to suppose that the propagation of gravitation is not instantaneous, but happens with the speed of light. This seems at odds with results obtained by Laplace, who announced that this propagation is, if not instantaneous, at least much faster than that of light. But in reality, the question posed by Laplace differs considerably from that which occupies us here. The introduction of a finite velocity of propagation was the only change Laplace introduced to Newton's law. Here, on the contrary, this change is accompanied by several others; it is possible, and that is indeed what happens, that a partial compensation occurs between them.
When we therefore speak of the position or velocity of the attracting body, it will be the position or the velocity at the time when the gravitational wave leaves the body; when we talk about the position or velocity of the attracted body, it will be the position or the velocity at the moment when this body was reached and attracted by the gravitational wave emanating from the other body; it is clear that the first instant precedes the second.
Poincare proved that in a relativistic theory of gravity, the forces can propagate as gravitational waves at the speed of light, and we would still have nearly elliptical orbits matching astronomical observations.
Poincare was completely correct. He was a decade ahead of everyone else.
To the relativity pioneers, Maxwell, Lorentz, and Poincare, local causality is at the core of the theory. Finding a way to resolve Laplace's paradox was crucial to establishing a causal and relativistic view of nature.
I am not sure that Einstein ever had good understanding of relativistic causality. He kept flip-flopping on gravitational waves, and on causality in quantum mechanics.
A couple of weeks after Einstein's library received this paper, Einstein submitted his own relativity paper, which was primarily a recapitulation of Lorentz's electromagnetic theory.
Einstein historian Galina Weinstein just posted:
Around 1936, Einstein wrote to his close friend Max Born telling him that, together with Nathan Rosen, he had arrived at the interesting result that gravitational waves did not exist, though they had been assumed a certainty to the first approximation. He finally had found a mistake in his 1936 paper with Rosen and believed that gravitational waves do exist. However, in 1938, Einstein again obtained the result that there could be no gravitational waves!Einstein did not find the mistake himself. He submitted it to a journal and a referee explained the error, as Lumo explains.
Tuesday, February 16, 2016
While the scientists were disappointed that the discovery was not real, the success of the analysis was a compelling demonstration of the collaboration's readiness to detect gravitational waves.And that Nature reports that "only three people would know the truth".
LIGO is famous for having papers coauthored by 100s of physicists. This 2015 paper has about 900 authors. That should make them more convincing, as 100s of physics could not all be lying. Usual scientific progress involves an openness to examining the raw data, but somehow they have engineered LIGO from the ground up for clandestine manipulation of the data.
But it is shocking that 3 people could have faked this result. I very much doubt that 3 people could have faked the Higgs boson discovery.
Who would set up a system for faking astrophysics with a "data injection"? And who brags about "compelling demonstration of the collaboration's readiness"? They spent a billion dollars on this detector, so obviously they have a readiness and willingness to report that they detected something.
When the Pope argued with Galileo over heliocentrism, did either of them use fake data injections? Do the global warming researchers use fake data injections in order to prove their readiness to accept a warming trend? Do the SETI folks broadcast faked space alien messages? The whole process is bizarre and fishy.
With the system engineered for fake data injection, there is also the possibility of a mistaken injection, and we may never know whether the data was legitimate or not.
Maybe the possibility of LIGO publishing fake data is no more likely than Antonin Scalia being murdered. But I think that possibility should be checked out also. It is the most ideologically convenient death in the history of the court, and there are many suspicious aspects to the story, and Scalia was expected to be the deciding vote in 6 big pending cases that are crucial to the leftist agenda.
Also, I got this correction:
Really, really insignificant correction: in the physics lingo (which does not necessarily constrain you) gravity waves are waves in some medium due to gravity, whereas what was discovered for the first time this week is gravitational waves.I am not sure about this distinction. The newly discovered gravitational waves are described as "ripples in the fabric of space–time". That sounds like waves in some medium due to gravity. But I guess "medium" means some medium other than space-time. This seems to be an old bugaboo, as some people say that light has no medium, while others say that the medium is the aether, space-time, or the QED vacuum.
Monday, February 15, 2016
Lawrence M. Krauss, a theoretical physicist at Arizona State University, walked us through four of Einstein’s notable blunders.I am actually going to defend Einstein here. He was right that spooky action at a distance is unphysical.
1. Quantum Entanglement
Einstein referred to this physical phenomenon, which suggests that objects separated by great distances can affect one another, as “spooky action at a distance.” He rejected the possibility, refusing to believe that objects could influence each other no matter how far apart they were.
“He didn’t think the spooky action at a distance would be verified, but it was,” Dr. Krauss said. “He thought that was somehow unphysical. He presented this as an example of why quantum mechanics is probably wrong, but in fact it’s right.” ...
[another physicist said] “Now we have confirmed that there is spooky action at distance.”
The quantum mechanics experiments show that observing one electron can affect our best prediction for another electron. And they are consistent with observations making an instaneous effect to the wave function. But there is no proof that an action on one electron can have an immediate physical effect on a distant electron.
It is not that complicated.
People have their beliefs in reality or ontology of the wave function, parallel universes, ideal point particles, Bayesianism, and all sorts of other things. Depending on such beliefs, they may or may not believe in spookiness. But Krauss and the other physicist are dead wrong when they say that spooky action at a distance was confirmed. It has not. No Nobel prize has ever been given for it, and textbook quantum mechanics does not require it.
I look forward to the day when today's physicists are mocked for all their silly beliefs that were proved dead wrong. Spooky action at a distance. Many worlds interpretation. Supersymmetry. Unified field theory. Black hole firewalls. Proton decay. Magnetic monopoles. Quantum cryptography. Quantum computing. 10-dimensional strings. Black hole holographic universes. Boltzmann brains. Tegmark multiverses. Entropic gravity.
Krauss also has a NY Times article explaining LIGO:
Ultimately, by exploring processes near the event horizon, or by observing gravitational waves from the early universe, we may learn more about the beginning of the universe itself, or even the possible existence of other universes.See also Michio Kaku in the WSJ:
This may also have philosophical implications. Right now the big-bang theory doesn’t tell us what banged, why it banged, and what caused it to bang. It only tells us that there was a bang. But if space-based gravity-wave detectors similar to LIGO’s detectors can measure the radiation emitted an instant after the big bang, then, using mathematics, one can run the equations backward to determine what set off the big bang in the first place, in effect answering the biggest question of all: What banged and why?
Sunday, February 14, 2016
LIGO is just claiming the observation of one event, so far. And that occurred very soon after upgrading to more sensitive equipment.
Media coverage of LIGO is remarkably similar to that generated by a previous gravitational-wave announcement. In March 2014, a team overseeing the Background Imaging of Cosmic Extragalactic Polarization observatory, or BICEP2, claimed to have detected gravitational waves produced by inflation, an extremely rapid--and hypothetical—cosmic growth spurt.
The lead BICEP2 researcher, John Kovac, assured The New York Times that “the chance that the results were a fluke was only one in 10 million.” I expressed doubts, saying I wanted “an explanation of why only inflation, and not other more conventional physical phenomena, can account for the gravity-wave findings.” Early in 2015, the BICEP2 researchers withdrew their claim, acknowledging that their observations had been distorted by dust in the Milky Way.
Now it is 5 months later. Has anything happened since? They are not telling us.
LIGO has cost American taxpayers about $1.1 billion. That is how much the National Science Foundation has spent on the project over the past 40 years, according to the Times.Chemists must be frustrated that esoteric physics gets all the money and publicity.
Chemist Ashutosh Jogalekar, who blogs as Curious Wavefunction, notes that while “the detection of gravitational waves will be a fitting testament to both experimental and theoretical science and the dedication of countless scientists over the years, in one sense it would be utterly unsurprising. That's because it is the logical prediction of a theory that has been around for a hundred years.”
Jogalekar adds that “some sources are already calling the putative finding one of the most important discoveries in physics of the last few decades. Let me not mince words here: if that is indeed the case, then physics is in bad shape.”
In an email to me, a technology scholar was more blunt: “So a 100 year old theory has been confirmed experimentally -- big whup. Did anyone think Einstein was wrong? There wasn't any controversy, was there? Was anyone credible claiming that spacetime isn't curved, or that black holes don't exist?
If confirmed by subsequent events, this result does plug a gap in our knowledge. We have found star-sized black holes, and giant million-star sized black holes at the nucleus of galaxies. But we had no evidence of medium-sized black holes, or of how they get so big. If the LIGO folks are right, then the black holes get big from collisions emitting detectable gravity waves.
My hunch is that they have not really nailed down the cause of the gravity wave, or told us what they know about other possible events. But we should soon see, as more data and analysis come in.
Friday, February 12, 2016
I previously attacked Richard Dawkins for wanting to fire someone for his personal beliefs, but he is a reasonable man compared to the other leftist-atheist-evolutionists.
Philosopher of pseudoscience Massimo Pigliucci writes about the latest controversy:
If you are following at all the skeptic / atheist / humanist / freethought movement(s) (henceforth, SAHF), last week has been an exciting and/or troubling one for you. First, the announcement that the Richard Dawkins Foundation had merged with (or taken over, depending on whom you ask) the venerable Center for Inquiry, up until then the chief remaining operation established by one of the founding fathers of modern skepticism and humanism, Paul Kurtz.To me, this just seems like the Stalinists purging the Trotskyists. Leftists cannot stand one of their fellow leftists deviating from their political dogma.
Then, a mere six days later, the organizers of the North East Conference on Science and Skepticism (NECSS), likely to soon become the major skeptic conference in North America (given the apparent demise of The Amazing Meeting), dropped a bombshell: Dawkins was being disinvited — probably a first in his career — on grounds of yet another obnoxious tweet he had thoughtlessly sent out to his 1.35 million followers. ...
The video linked to in the tweet, and which Dawkins clearly endorsed, can be found here. It is an egregious, unqualified, piece of racist and misogynist garbage. Please, pause reading this post for a couple of minutes and see for yourself. It’s simply horrifying.
Then again, this was not an isolated incident. Dawkins had racked a considerable number of similarly embarrassing tweets over the past few years. Here is a sampler, ranging over such light topics as abortion, rape, pedophilia, and Islam (of course!). Use Google to find many, many more.
Pigliucci cries "racist", but denies that there is any such thing as race. The video makes no mention of race. I can only assume that he is projecting his own racial prejudices.
It is baffling how these folks can call themselves the "freethought movement", and then spend most of their time banning each other for being divisive.
The video brilliantly mocks Islamism and feminism, and their not-so-obvious similarities. Yes, it distorts their beliefs somewhat, I guess.
A reader writes:
As for scientists denying free will, I concur in part. Many of them have no free will whatsoever, as they are prisoners of their own groupthink, confusing political consensus with scientific comprehension.Do these people really not have the ability to watch a silly satirical video and think for themselves?
If they say that they do not have the ability, maybe I should take their word for it. They want an enforced groupthink, where no one dares go against the hivemind.
Twitter has just announced that it will more actively block tweets like this:
With hundreds of millions of Tweets sent per day, the volume of content on Twitter is massive, which makes it extraordinarily complex to strike the right balance between fighting abuse and speaking truth to power. It requires a multi-layered approach where each of our 320 million users has a part to play, as do the community of experts working for safety and free expression.Those organizations and experts are mostly leftist social justice warriors who believe that safety means protection from exposure to offensive political opinions.
That’s why we are announcing the formation of the Twitter Trust & Safety Council, a new and foundational part of our strategy to ensure that people feel safe expressing themselves on Twitter. ... We have more than 40 organizations and experts from 13 regions joining as inaugural members of the Council.
Twitter is a big disappointment. So is Facebook. The apps are obnoxious and the companies are run by creeps. The public never should have let one company control everyone's web page, and another company control everyone's public comments. Couldn't anyone devise a more open system, like email?
Actually email is being taken over by Microsoft, Google, and Yahoo. They are the only ones with the resources and know-how to control spam. At least we have 3 choices, until Yahoo goes bankrupt.
Update: Dawkins has suffered a damaging stroke. Pigliucci backpeddles, but read the comments where he has to be corrected on several points.
Thursday, February 11, 2016
A team of physicists who can now count themselves as astronomers announced on Thursday that they had heard and recorded the sound of two black holes colliding a billion light-years away, a fleeting chirp that fulfilled the last prophecy of Einstein’s general theory of relativity.If correct, this is indeed a big astrophysics advance. About a billion dollars in obscure research has a positive result.
That faint rising tone, physicists say, is the first direct evidence of gravitational waves, the ripples in the fabric of space-time that Einstein predicted a century ago (Listen to it here.). And it is a ringing (pun intended) confirmation of the nature of black holes, the bottomless gravitational pits from which not even light can escape, which were the most foreboding (and unwelcome) part of his theory.
More generally, it means that scientists have finally tapped into the deepest register of physical reality, where the weirdest and wildest implications of Einstein’s universe become manifest.
Conveyed by these gravitational waves, an energy 50 times greater than that of all the stars in the universe put together vibrated a pair of L-shaped antennas in Washington State and Louisiana known as LIGO on Sept. 14.
I have just a couple of observations, as I have not read the papers.
The biggest physics a announcement of last year was the BICEP2, which was supposed to be the echo of gravity waves during the big bang. It got glowing endorsements from everyone, as proof of big bang, inflation, quantum gravity, and gravity waves. Only months later did a rival group convince everyone that they just saw some boring cosmic dust.
Next, all the talk of Einstein is a little silly. He did not even believe in gravity waves. He wrote a paper in 1936 that supposedly disproved gravity waves, and then got angry when an anonymous referee proved him wrong. He thought that he was too good to be subjected to the referee process.
Third, the LIGO experiment is an interferometer that is quite similar to the Michelson-Morley experiment that was the crucial experiment for special relativity. The MMX was trying to detect the motion of the Earth, while LIGO is trying to detect motion in another galaxy.
Wednesday, February 10, 2016
Back in the 1980s, the American scientist Benjamin Libet made a surprising discovery that appeared to rock the foundations of what it means to be human. He recorded people’s brain waves as they made spontaneous finger movements while looking at a clock, with the participants telling researchers the time at which they decided to waggle their fingers. Libet’s revolutionary finding was that the timing of these conscious decisions was consistently preceded by several hundred milliseconds of background preparatory brain activity (known technically as “the readiness potential”).No, it just meant that it can take a few seconds for the human brain to make a decision. Not surprising, and not relevant to free will.
The implication was that the decision to move was made nonconsciously, and that the subjective feeling of having made this decision is tagged on afterward. In other words, the results implied that free will as we know it is an illusion — after all, how can our conscious decisions be truly free if they come after the brain has already started preparing for them?
Now, some more precise research has been done, and the conclusion is the opposite of Libet's:
For years, various research teams have tried to pick holes in Libet’s original research. ...When scientists tell you something contrary to common sense, such as experiments proving that you have no free will, you should consider the possibility that they are completely wrong.
These studies all point in the same, troubling direction: We don’t really have free will. In fact, until recently, many neuroscientists would have said any decision you made was not truly free but actually determined by neural processes outside of your conscious control.
Luckily, for those who find this state of affairs philosophically (or existentially) perplexing, things are starting to look up. Thanks to some new breakthrough studies, including one published last month in Proceedings of the National Academy of Sciences by researchers in Germany, there’s now some evidence pointing in the other direction: The neuroscientists are backtracking on past bold claims and painting a rather more appealing account of human autonomy. We may have more control over certain processes than those initial experiments indicated.
Monday, February 8, 2016
In the libertarian ``agent causation'' view of free will, free choices are attributable only to the choosing agent, as opposed to a specific cause or causes outside the agent. An often-repeated claim in the philosophical literature on free will is that agent causation necessarily implies lawlessness, and is therefore ``antiscientific." That claim is critiqued and it is argued, on the contrary, that the volitional powers of a free agent need not be viewed as anomic, specifically with regard to the quantum statistical law (the Born Rule). Assumptions about the role and nature of causation, taken as bearing on volitional agency, are examined and found inadequate to the task. Finally, it is suggested that quantum theory may constitute precisely the sort of theory required for a nomic grounding of libertarian free will. ...This may sound ridiculous, but quantum mechanics allows for free will in a way that no other physical theory does. It predicts many things while leaving others unpredictable. Maybe that unpredictability is part of a free will mechanism.
Considering the elementary constituents of matter as imbued with even the minutest propensity for volition would, at least in principle, allow the possibility of a natural emergence of increasingly efficacious agent volition as the organisms composed by them became more complex, culminating in a human being. And allowing for volitional causal agency to enter, in principle, at the quantum level would resolve a very puzzling aspect of the indeterminacy of the quantum laws–the seeming violation of Curie’s Principle in which an outcome occurs for no reason at all. This suggests that, rather than bearing against free will, the quantum laws could be the ideal nomic setting for agent-causal free will.
You may not accept this, but it does disprove the idea that free will is anti-scientific. Many scientists and most philosophers argue that a scientific outlook requires denying free will. They are wrong. Quantum mechanics allows free will.
Thursday, February 4, 2016
Horgan: Nice! You recently said on your blog: “The biggest task of science bloggers -- like Peter Woit, Ethan Siegel and myself -- has become to clean up after sloppy science journalism.” Please elaborate.She does not mention Lubos Motl, as he is probably the leading physics blogger cleaning up sloppy journalism, even if he is also the most annoying.
Hossenfelder: I often find myself having to correct articles that mislead the reader about some recent research. The way much science journalism appears today, it is impossible for someone with no background in the field to tell how serious to take claims. Like, that new research shows black holes don’t exist, or that we will make contact with parallel universes, will soon test quantum gravity, or string theory, or that the information loss paradox has been solved (again!). And so on.
People don’t learn from this, they just get confused, doubt the trustworthiness of science, and it’s no good. I recently went to visit my mom and first thing she says after she opens the door is that she’s read the LHC proved we live in a multiverse and if I could please tell her what that is supposed to mean.
I have criticized her several times, most recently on free will.
I don't really attack sloppy science journalism much. I can hardly blames the journalists for taking the wacky ideas from big-shot physicists seriously. I blame the big-shot physicists for spewing nonsense. And I also blame other scientists and philosophers with high-status jobs.
Wednesday, February 3, 2016
The perceptron was a brilliant advance in the 1950s. In the last 30 years it has been improved to solve all sorts of problems. But in 1969, Minsky convinced everyone that the field was a dead-end, and progress stopped in that direction.
I am not trying to put Minsky. A lot of smart people misjudged the potential of perceptrons.
A 1972 book review said:
This book is a very interesting and penetrating study of the power ofWow, I would have said that dozens of XXc math theories were way more important. But related ideas are now being used for automatic face recognition, self-driving cars, voice commands, robotics, etc. Maybe someone will consider this a huge advance to enable our robot overlords to enslave us.
expression of perceptrons and some other mathematical problems concerning memory and learning. This subject is still quite new and hence at a stage of development in which the most important discoveries are being done. It seems to differ from the theory of automata in its greater relevance to our ideas about the organization of the brain and the construction of "such" machines. ... It is also the first purely mathematical monograph about certain aspects of learning and perception, and this subject may become the most important theory of 20th century mathematics.
Tuesday, February 2, 2016
It is 100 years since the publication of Einstein's great theory, and arguably one of the greatest scientific theories of all time. To mark the occasion, Brian Cox takes Robin Ince on a guided tour of General Relativity. With the help of some of the world's leading cosmologists, and a comedian or two, they explore the notions of space time, falling elevators, trampolines and bowling balls, and what was wrong with Newton's apple. It is a whistle stop tour of all you will ever need to know about gravity and how a mathematical equation written 100 years ago predicted everything from black holes to the Big Bang, to our expanding universe, long before there was any proof that these extraordinary phenomena existed.They try to give some short simplified descriptions of general relativity, but they seem unsatisfactory to me. So I will try.
General relativity is a way of reconciling gravity with the causal structure of spacetime. Newtonian gravity involves masses exerting a force over a large distance, without any intervening mechanism. Maxwell devised a causal theory of electromagnetism, where charges generate local fields that propagate to interact with other charges. Relativity interprets this in terms of a geometry of space and time, and describes gravity as a similar curvature field that propagates to other masses.
They said that GR was developed from theory, not experiment. That is partially true, but I would not say it that way. Special relativity was driven by trying to reconcile the Michelson-Morley experiment with Maxwell's equations. The result was spacetime non-Euclidean geometry. Was that theory or experiment? Some of each.
A commenter notes that Einstein's work did not directly use the Michelson-Morley experiment. That is true, but even Einstein conceded that the experiment was crucial for the discovery of special relativity. It just was not needed for Einstein's recapitulation of Lorentz's work.
Solar system observations were the experimental data for gravitational theory. Newton and Laplace showed that gravity must be instantaneous or propagate much faster than light to explain planetary orbits. Poincare used relativity to show that gravity could propagate at the speed of light, and also to partially explain anomalies in Mercury's orbit. Einstein showed that a Ricci-flat spacetime could fully explain the Mercury anomaly, and that convinced him that he had the correct relativistic gravity theory.
Other aspects of GR were not properly tested until decades later, but experiment was still crucial for its original development.
In the last month there have been a lot of rumors about a pending announcement of the discovery of gravity waves, and the news stories always say that this would be a vindication of Einstein's theory of general relativity. As a reader points out, Einstein himself did not believe in gravity waves, and tried to publish a paper proving that they did not exist. It was Einstein's only refereed paper, and the referee said that it was wrong.
Poincare proposed a theory of relativistic gravity waves in 1905. He was decades ahead of Einstein on this point.