Seth Lloyd - What Can't Be Predicted in Physics?

Prediction is the fruitful product of good science, but how far can prediction go? Physics is the most mathematical and rigorous of the sciences and so prediction is most successful in physics. But are there limits to predictability in physics? What about quantum indeterminacy? Are there ultimate barriers to prediction in physics?



Seth Lloyd is a professor of mechanical engineering at the Massachusetts Institute of Technology. He refers to himself as a "quantum mechanic".


Closer to Truth, hosted by Robert Lawrence Kuhn and directed by Peter Getzels, presents the world’s greatest thinkers exploring humanity’s deepest questions. Discover fundamental issues of existence. Engage new and diverse ways of thinking. Appreciate intense debates. Share your own opinions. Seek your own answers.

[shiddy.]

good job with the audio on this one

that was brave to continue the interview with the wind - the gamble paid off

[melgross]

Seth is one of my favorite interviewees.

[friendlystonepeople]

This information theory viewpoint of the universe is extremely mind blowing.

[wayneasiam]

Great Seth Lloyd interview. Closer to Truth. Robert Kuhn is good at getting very smart people to open their minds to us. Keep striving, Humanity! We're getting there. Who knows when the greatest discovery of all is about to happen. Then we'll take a quantum leap in physics, biology, all of it. We're in a precipice golden age for the advancement of everything.

[Yzjoshuwave]

Seth is hitting a bunch of really interesting notes here - I like his energy and his approach. His position still hinges on the assumption that the physics of a transcendental computer capable of running, or modeling the Universe will have the same constraints as the physics of the Universe, which is dubious. If the laws of the Universe are understood to be “programmed in”, we might need to consider a much larger category-space of possible computation.

[-by-_Publishing_LLC]

(8:55) *SL: **_"No matter how many regularities are built into the world, you still won't be able to predict stuff."_* ... The _predictability_ associated with science and physics is great whenever one is dealing with things that are subject to predictability. However, "Existence" requires an oppositional reference point for everything in existence. That means wherever you find _predictability, _ you'll also find a compensational degree of _unpredictability._

Not everything in Existence can be explained via science, physics, and quantum theory. You need other less-restrictive methods of information processing to deal with the unpredictable.

At the end of the day, _chaos_ exists in equal proportion to _order._ ... Otherwise there is no comprehension of either.

[WilliamJohnston]

No joke - I was thinking about simulations of the universe recently and tried to explain to my wife my hypothesis that a computer within the universe would never be able to simulate the universe on a one-to-one scale faster than the universe is already running… I was so glad to see this video!

[jjcm]

brilliant conversation by 2 enjoyable men.

[david_porthouse]

Behaviour which violates Bell’s inequalities would appear to be random enough to be useful for the construction of a reliable Vernam cipher, and even more random than that: every time we do a Lorentz transformation we have a new random sequence to deal with. For the purpose of a computer simulation of the Solar System, this randomness can be modelled as classical Lucretian motion on the scale of Planck’s constant. We then run the simulation a few times to get an idea of typical behaviour.

[larrysherk]

You are free to predict anything you like in physics. The problem is discovering an acceptable probability of success. It's never very high unless it's trivial.

[David.C.Velasquez]

Here's a thought. Consider a CD, the media type is actually of no concern and only an example, but just by starting at "all zeroes" then count those 780MB of bits up to all ones... in between exists every album ever recorde, and that could ever be recorded. Also every possible data or string of images, etc. At least, any data conceivable within a given encode/decode space, I would assume.

[bruinflight]

Being able to unravel and discern a piece of what happened in the past is too often taken as an ability to predict what will happen in the future

[philosophyetc]

An fascinating lecture related to this is Stephen Hawking's "Gödel and the End of Physics." The transcript is available online.

[longcastle]

I think determinism will eventually be found to be an illusion -- or at the very least be understood as a human idea or description of some system of occurrences in the universe that eventually breaks down within certain kinds of models or at certain levels of scrutiny...

[JungleJargon]

The universe is not a computer and it is very regular inspite of what we still don't know about it. The laws of physics are the same everywhere except for what we don't know about inextreme conditions both with energy and matter and without energy and matter.

[ZenRyoku]

the only things that cannot be explained by physics is non-tangible abstract notions...💯

[julianmann]

Newton pointed out that although the motions of planets can be predicted using his formulae, the initial state of the solar system can't be and we know that gravity travels at C. So for instance it takes 8 minutes for the Sun's gravity to reach Earth. So this is an example where physics can't predict everything. Also we know that time has two arrows, which will undermine deterministic predictions at the quantum level.

[mykrahmaan]

Prediction need not be the purpose of knowledge. On the contrary, knowledge should only be searched to achieve a desirable future state. Once this desirable state is formulated, one must compare what currently exists with it to derive what changes must be made to what exists so that the desired state is achieved.

One doesn't go to a mall to decide what to purchase, but to buy what one already has decided to acquire.

Similarly, if we restrain our search for knowledge by what we intend to do with it this question of predictability doesn't arise at all.

The quantum uncertainty indicates that we MUST restrain our search for knowledge by our needs and the imaginations be tethered to the means for satisfying so defined needs.

That way all these purpose free speculations would vanish.

[jamesruscheinski]

Better understanding of causation needed to predict the future?

[davidrandell]

D=1/2 at^2 real predictive of Galilean relative. The earth approaches the released object. 50% chance as compared to the object ‘falling ‘ or ‘dropping ‘. Real simple. Juliana Mortenson website Forgotten Physics classicalized QM in 2010. Try to keep up.