scott aaronson associate professor, eecs
DESCRIPTION
The Limits of Computation. Quantum Computers and Beyond. Scott Aaronson Associate Professor, EECS. Moore’s Law. Extrapolating: Robot uprising?. But even a killer robot would still be “merely” a Turing machine, operating on principles laid down in the 1930s…. =. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Scott Aaronson Associate Professor, EECS](https://reader038.vdocuments.net/reader038/viewer/2022103008/568167f3550346895ddd6839/html5/thumbnails/1.jpg)
Scott AaronsonAssociate Professor, EECS
Quantum Computers and Beyond
![Page 2: Scott Aaronson Associate Professor, EECS](https://reader038.vdocuments.net/reader038/viewer/2022103008/568167f3550346895ddd6839/html5/thumbnails/2.jpg)
Moore’s Law
![Page 3: Scott Aaronson Associate Professor, EECS](https://reader038.vdocuments.net/reader038/viewer/2022103008/568167f3550346895ddd6839/html5/thumbnails/3.jpg)
Extrapolating: Robot uprising?
![Page 4: Scott Aaronson Associate Professor, EECS](https://reader038.vdocuments.net/reader038/viewer/2022103008/568167f3550346895ddd6839/html5/thumbnails/4.jpg)
But even a killer robot would still be “merely” a Turing machine, operating on
principles laid down in the 1930s…
=
![Page 5: Scott Aaronson Associate Professor, EECS](https://reader038.vdocuments.net/reader038/viewer/2022103008/568167f3550346895ddd6839/html5/thumbnails/5.jpg)
Is there any feasible way to solve these problems, consistent with
the laws of physics?
And it’s conjectured that thousands of interesting problems are inherently
intractable for Turing machines…
![Page 6: Scott Aaronson Associate Professor, EECS](https://reader038.vdocuments.net/reader038/viewer/2022103008/568167f3550346895ddd6839/html5/thumbnails/6.jpg)
Relativity Computer
DONE
![Page 7: Scott Aaronson Associate Professor, EECS](https://reader038.vdocuments.net/reader038/viewer/2022103008/568167f3550346895ddd6839/html5/thumbnails/7.jpg)
Zeno’s Computer
STEP 1
STEP 2
STEP 3STEP 4
STEP 5
Tim
e (s
econ
ds)
![Page 8: Scott Aaronson Associate Professor, EECS](https://reader038.vdocuments.net/reader038/viewer/2022103008/568167f3550346895ddd6839/html5/thumbnails/8.jpg)
Time Travel Computer
R CTC R CR
C
0 0 0
Answer
“Causality-Respecting Register”
“Closed Timelike
Curve Register”
Polynomial Size Circuit
S. Aaronson and J. Watrous. Closed Timelike Curves Make Quantum and Classical Computing Equivalent, Proceedings of the Royal Society A 465:631-647, 2009. arXiv:0808.2669.
![Page 9: Scott Aaronson Associate Professor, EECS](https://reader038.vdocuments.net/reader038/viewer/2022103008/568167f3550346895ddd6839/html5/thumbnails/9.jpg)
What we’ve learned from quantum computers so far:
15 = 3 × 5(with high probability)
![Page 10: Scott Aaronson Associate Professor, EECS](https://reader038.vdocuments.net/reader038/viewer/2022103008/568167f3550346895ddd6839/html5/thumbnails/10.jpg)
Linear-Optical Quantum Computingwww.scottaaronson.com/papers/optics.pdf
My student Alex Arkhipov and I recently proposed an experiment, which involves generating n identical photons, passing them through a network of beamsplitters, then measuring where they end up
Our proposal almost certainly wouldn’t yield a universal quantum computer—and indeed, it seems a lot easier to implement
Nevertheless, we give complexity-theoretic evidence that our experiment would solve some sampling problem that’s classically intractable
Groups in Brisbane, Australia and Imperial College London are currently working to implement our experiment
![Page 11: Scott Aaronson Associate Professor, EECS](https://reader038.vdocuments.net/reader038/viewer/2022103008/568167f3550346895ddd6839/html5/thumbnails/11.jpg)
Summary1. From a theoretical standpoint, modern
computers are “all the same slop”: polynomial-time Turing machines
2. We can imagine computers that vastly exceed those (by using closed timelike curves, etc.)
3. But going even a tiny bit beyond polynomial-time Turing machines (say, with linear-optical quantum computers) is a great experimental challenge