Clojurians Log v2

I've been reading a lot about people who've been complaining about how SaaS and App vendors have been changing interfaces on them in bad ways and I'm always glad that I'm in control of what changes in emacs

I think one time magit changed the UI...and it was tiny and I liked it. Sums it up I think

GNU Morning

I know nothing about quantum computing but there is a https://multiversecomputing.com/ near us compressing AI models with quantum algorithms, and yesterday a quantum computer from IBM arrived https://www.diariovasco.com/gipuzkoa/entra-superordenador-cuantico-ibm-20251015235704-nt.html#vtm_modulosEngag=historias-visuales:portada:noticia:1 too.

I really think QC is getting hot now. And the field is still small enough to get a grip on, if you start out learning.

Compacting AI models sounds interesting.

I think they pivoted recently, the company was founded in 2019 and they were selling quantum algorithms for other use cases.

So, what's quantum about your quantum circuit? What's gained that isn't gained through regular computing?

@lsolbach where is a good place to start w/quantum computing? It feels like the kind of thing that might help some of our simulation work that we currently do, but I wonder if I've completely got the wrong end of the stick with regards to applications

@otfrom If you want to use Clojure, take a look at https://cljdoc.org/d/org.soulspace/qclojure/0.21.0/doc/tutorial

A good book also helps, e.g. Dancing with Qubits

@reefersleep in principle, you can do the same things with quantum or with classical computing. But some problems can be calculated faster on quantum computers than on classical computers. Speedups range from quadratic (search in unstructured data) up to exponential (e.g. prime factorization). But it is still a reseach area.

@lsolbach very nice 🤓 Is your usage a particular application or general research?

(or maybe it's under NDA 🕵️‍♂️)

I've been exposed to quantum computing this spring and I wanted to gain expertise on it. And most existing frameworks are in Python. So I combined building a Clojure solution with my learning.

So it was completely self motivated?

Yes, kind of.

Amazing

Never underestimate intrinsic motivation. 😄

I give two talks at the macroexpand conferences: "Functional Quantum Computing with QClojure" tomorrow and "Building QClojure with AI Assistance" next friday.

Was renting the quantum hardware expensive? (I assume that's what you did)

I have to check the bill. It's not too expensive with Amazon Braket, I would say. Playing around is affordable. But it depends a lot on what you are trying to do. One problem with quantum computing is it's probabilistic nature, so you have to run your circuits for a number of times to get meaningful results. The other problem is the state of the current hardware, which is noisy. You can cope by either run even more shots or by blowing up your circuits with error correction, which can be prohibitive, depending on the actual hardware.

Sounds so strange and foreign 🤓