I've been bit by the mass marketing nonsense of "Watson" but IBM Research does some pretty good work, and their progress on Quantum Computing seems to be "real"; and certainly more reliable than Microsoft (shocked!).
> IBM anticipates that the first cases of verified quantum advantage will be confirmed by the wider community by the end of 2026.
In 2019, Google claimed quantum supremacy [1]. I'm truly confused about what quantum computing can do today, or what it's likely to be able to do in the next decade.
A decade from now Quantum computing will be in the same place it was a decade ago, on the cusp of proving a quantum advantage for tailor made problems in comparison to normal availability supercomputers. Classical compute will advance in that time period to keep the quantum computers always on the cusp.
The major non-compute related engineering breakthroughs needed for quantum computing to actually be advantageous in a way that would be revolutionary are themselves so revolutionary that the advancements of quantum computing would be vastly overshadowed. Again it's a case where those breakthroughs would so greatly enhance classic compute in terms of processing and reduction in costs that it still probably wouldn't be economically viable to produce general purpose quantum computers.
"Qiskit capabilities show 24 percent increase in accuracy"
what was it before? What good is a computer that is not 100% accurate? Do I have to run a function 1000x to get some average 99% chance the output is correct?
Related Qiskit Tutorial Video[0]
"This tutorial covers advanced techniques for implementing the Quantum Approximate Optimization Algorithm (QAOA) at the utility scale using Qiskit. In this video, we walk through how to build, optimize, and run QAOA for real world optimization problems on real IBM Quantum hardware.
This series is designed for quantum computing practitioners who are ready to move beyond basic examples and start running large scale, hardware aware algorithms. We explore how to transition from theory to practical execution, covering algorithm development, circuit optimization, hybrid workflows, and best practices for hardware performance. Whether you are expanding your QAOA skills or preparing to run your own research experiments, this tutorial will help you strengthen your understanding of utility scale quantum computing with Qiskit."
I happen to know IBM made some great hires -- one of my classmates who was excellent in the field, who had impressive quantum computing nature publications before graduation, worked at IBM for the past several years.
Though it looks like he recently switched to working at Google AI...
Anyone getting use of their money via Red-Hat sponsored projects like Linux kernel, GNOME and GCC, OpenJDK, Quarkus, VSCode plugins for Java for example.
11 comments
[ 1.4 ms ] story [ 34.9 ms ] threadIn 2019, Google claimed quantum supremacy [1]. I'm truly confused about what quantum computing can do today, or what it's likely to be able to do in the next decade.
[1] https://www.nasa.gov/technology/computing/google-and-nasa-ac...
The major non-compute related engineering breakthroughs needed for quantum computing to actually be advantageous in a way that would be revolutionary are themselves so revolutionary that the advancements of quantum computing would be vastly overshadowed. Again it's a case where those breakthroughs would so greatly enhance classic compute in terms of processing and reduction in costs that it still probably wouldn't be economically viable to produce general purpose quantum computers.
[0] https://www.youtube.com/watch?v=rBfK-l-qSNk
Though it looks like he recently switched to working at Google AI...
https://scholar.google.com/citations?user=NaxMJzQAAAAJ&hl=en
I wonder what would happen to them if codex or what have helps migrate that to c#.
= how long until the exodus to aws/azure will follow