@chatsagnik.bsky.social
Quantum Computing, Learning Theory, Boolean functions. PhD@IIITD. Website: https://chatsagnik.github.io/ Blog: https://theoretickles.netlify.app/
Great paper! If you are free sometime, let me know. I would love to chat!
17.10.2025 06:57 — 👍 0 🔁 0 💬 0 📌 0
Today on X: Came across a post by someone I deeply respected as a researcher.
The context in this post is illegal immigration to "first world countries".
I mean wtf?
I get that people in the "Global North" don't want illegal immigration. That's fair.
WTF though?
This sounds like a nightmare scenario. Maybe escalating the matter up the editorial chain would help.
05.07.2025 06:45 — 👍 0 🔁 0 💬 0 📌 01. What about escalating the matter to higher ups at TMLR?
2. At all three conferences did the metareview go against the majority review sentiment? That's an outlier in itself!
P.S. Obligatory tagging of the meme curator @dulwichquantum.bsky.social
11.06.2025 18:37 — 👍 0 🔁 0 💬 0 📌 0
This is joint work with @yash-saxena.bsky.social and @tharrma.bsky.social.
Both of them did 99.99% of the heavy lifting on the paper😂 and are on the lookout for opportunities! (PhD and job resp.).
In the Bsky tradition, enjoy this meme!
P.S. Comments on the work welcome!
📜🚨Excited to report arxiv.org/abs/2504.15265!
We give a microwave-only, parametric framework (GCR) that directly enables entanglement on the 1–2 subspace (inaccessible via CR) and achieves gate fidelities up to 99.73%, alongside 99.06%-fidelity Bell state in 514ns (beating cross-Kerr SOTA).
I know for sure that similarly flavoured papers were not accepted for poster presentations at earlier editions of QIP. The very presence of this paper at TQC (!!) sends a very concerning signal to newcomers imo.
02.06.2025 09:43 — 👍 1 🔁 1 💬 0 📌 0I opened the paper to figure out exactly what they were sweeping under the rug. Hoo boy!
"We prove here that NP⊆BQP, by presenting a BQP-algorithm for the NP-complete variant of the SUBSET − SUM problem, without any restriction or approximation or assumption."
😂😂
Very nice result by @idanattias.bsky.social and coauthors!
arxiv.org/abs/2505.06146
The QED at the end really cements this analogy.
17.05.2025 07:29 — 👍 0 🔁 0 💬 0 📌 0
Excited to share my 5-lecture mini-course 🎬 on "Quantum Recursive Programming"! An elegant way to program complicated quantum algorithms ⚛️
*No prior QC or PL knowledge is needed!
Given during my visit to DIMACS at Rutgers University.
(www.youtube.com/playlist?lis...)
(1/4)
Today #quantum arXiv selection: quantum error correction (QEC) for trapped ions, quantum simulation of open quantum system, noise-specific QEC, variational quantum algorithm, quantum annealing with error mitigation, and data encoding.
More details and links below:
I suspected this was the case especially after a post titled "Time is running out!" xD
16.03.2025 17:44 — 👍 1 🔁 0 💬 0 📌 0"* bias"..... Learning theory is doomed.
10.03.2025 03:51 — 👍 2 🔁 1 💬 0 📌 0
Hooray, I received a Vici grant from the Dutch science foundation!
Heads up for current PhD students in learning theory: I will have two postdoc positions available in Amsterdam on "learning theory for interpretable/explainable AI" in the coming years.
www.uva.nl/shared-conte...
Why did I not fight back?
Short answer: My collaborator's PhD would be jeopardized.
Today I was kicked off a paper, I had conceived, led for a year, and done most of the lifting for.
Why? I objected to the fact that someone got added with no contribution, simply by virtue of being my collaborator's advisor.
One of those things that you cannot believe can happen; until it does.
Great work! Congrats @aangrisani.bsky.social
29.01.2025 05:25 — 👍 1 🔁 0 💬 0 📌 0This is a special paper for me:
- It is my 1st purely classical (and classical learning) result,
- It is my 2nd consecutive (and overall) AISTATS publication, and
- It is the 3rd (and final) published result in my thesis.
I look forward to meeting up with others in Phuket!
(3/3)
See arXiv:2405.13666 for an earlier draft (updated version coming soon).
A concurrent work (arXiv:2406.12600) by @neu-rips.bsky.social was recently accepted at ALT 2025, and addresses the same question!
The two papers use very different techniques to arrive at the same result! (2/3)
A good start to the new year!
📜Our paper "Generalization Bounds for Dependent Data using Online-to-Batch Conversion" was accepted to AISTATS 2025!
In this work, we give generalization bounds for statistical learners trained on non-i.i.d. data using the Online-to-Batch conversion framework. (1/3)
If your concern is wrt jobs, I believe that the jobs that can be automated should very well be.
From my time as an SWE, I observed that most companies bloat their payroll with underpaid employees so that they can overcharge clients.
If you are concerned about other stuff, I would love to hear it.
Congratulations! 🎉
23.12.2024 20:52 — 👍 1 🔁 0 💬 0 📌 0A video summarising the first time I tried opening a window in Europe.
22.12.2024 13:24 — 👍 3 🔁 0 💬 0 📌 0
*PLS SHARE*
Open position for an *Associate Professor* in Machine Learning at our department (@enginyeria-upf.bsky.social / @upf.edu), via the Serra Hunter programme.
DEADLINE: January 13th 2025
www.upf.edu/web/personal...
Announcing the accepted papers for ALT 2025! It was a very competitive year, and we were able to accept only 51 of 144 submissions. Thanks to my co-chair Po-Ling Loh and the entire program committee for their hard work!
See you in Milan in February!
algorithmiclearningtheory.org/alt2025/acce...
Making a #gamedev starterpack for (mostly) solo game developers working on cool projects. I've added a bunch of friends of mine you should all check out.
Leave a comment below if you're a solo #indiedev and want to be added!
go.bsky.app/15UnZcn
At QTML 2024, I spoke about recent work with Robert Huang on "Learning shallow quantum circuits with many-qubit gates" (a.k.a. efficient learning of QAC^0 unitaries). In this ~15min talk I discuss the project motivation, key results, and high-level proof ideas.
www.youtube.com/watch?v=iRiJ...
It seems like a tradition is emerging here, and it is out duty to maintain it. So here is my part announcing the publication of our review in semidefinite programming for characterizing quantum correlations @dulwichquantum.bsky.social
journals.aps.org/rmp/abstract...
