Maissam Barkeshli
@MBarkeshli
Followers
3K
Following
973
Media
41
Statuses
663
Visiting Researcher @ Meta FAIR. Professor of Physics @ University of Maryland & Joint Quantum Institute. Previously @ Berkeley,MIT,Stanford,Microsoft Station Q
University of Maryland, College Park
Joined December 2011
An absolutely incredible, highly interconnected web of ideas connecting some of the most important discoveries of late twentieth century physics and mathematics. This is an extremely abridged, biased history (1970-2010) with many truly ground-breaking works still not mentioned:.
10
128
486
I’m excited to be part of the new @SimonsFdn Simons Collaboration on the Physics of Learning and Neural Computation!.
Our new Simons Collaboration on the Physics of Learning and Neural Computation will employ and develop powerful tools from #physics, #math, computer science and theoretical #neuroscience to understand how large neural networks learn, compute, scale, reason and imagine:.
0
1
34
RT @dayal_kalra: 🤖 Transformers can write poetry, code, and generate stunning art, but can they predict seemingly random numbers? . We show….
0
5
0
well said @DimaKrotov.
Nice article! I appreciate that it mentions my work and the work of my students. I want to add to it. It is true that there is some inspiration from spin glasses, but Hopfield is much bigger than spin glasses. The key ideas that resurrected artificial neural networks in 1982.
0
0
3
RT @dayal_kalra: Excited to share our paper "Universal Sharpness Dynamics. " is accepted to #ICLR2025!. Neural net training exhibits rich….
0
59
0
RT @theeczoo: Chetan Nayak overviews unpublished results by @MSFTResearch on Majorana qubits.
0
19
0
Adam is one of the most popular optimization algorithms in deep learning but it is very memory intensive. We did some analysis to understand how much of the second moment information can be compressed, leading to SlimAdam. Thanks @dayal_kalra @jwkirchenbauer @tomgoldsteincs.
Low-memory optimizers sometimes match Adam but aren't as reliable, making practitioners reluctant to use them. We examine when Adam's second moments can be compressed during training. We also introduce SlimAdam, which compresses moments when feasible & preserves when detrimental
1
0
10
RT @SuryaGanguli: Where would we as a society be without math? It is a travesty that our entire public investment in math is ONLY $250M (0….
0
71
0
Transformers and pseudo-random numbers.We looked at the ability of transformers to learn arithmetic sequences obtained from simple pseudo-random number generators (PRNGs). The model spontaneously discovered prime factorization and residual number-system representations
3
4
41
If you want to read more about Tsuei and Kirtley's amazing experiment, check out this review article: . And here is our paper studying what is effectively the electric-magnetic dual of this phenomenon in topological insulators:.
arxiv.org
In the presence of crystalline symmetries, topological phases of matter acquire a host of invariants leading to non-trivial quantized responses. Here we study a particular invariant, the discrete...
0
0
2
Recently, we've understood a kind of electric-magnetic dual of this phenomenon. In certain kinds of insulators (an insulator is like the electric-magnetic dual of a superconductor), crystal defects can trap a quantized fractional amount of electric charge!
1
0
1
Tsuei and Kirtley actually measured this in 1994. They grew YBCO on a SrTiO3 substrate with a trijunction, and then used a scanning SQUID magnetometer to sensitively measure the local magnetic field.
0
0
2
Amazingly, at the trijunction, and only at the trijunction, there will be a trapped magnetic flux of exactly hc/4e (half a superconducting flux quantum).
1
0
3
Take a cuprate superconductor (in this case YBCO) and grow it on a tricrystal so that it forms a trijunction, like this:
1
0
2
A propos of nothing, here is one of the most beautiful experiments in physics from the last 30 years:
4
6
38
RT @SuryaGanguli: Overall, I believe it is imperative to pursue a new unified Science of Intelligence that can help us understand and impro….
0
1
0
We made some progress in the theory of quantum critical points enriched with crystalline symmetry. I'm happy to have teamed up with the great quantum field theorist @ZoharKo along with impressive students @CFechisin and Siwei Zhong.
What happens when fermions jump in a crystal with defects? With Chris, @MBarkeshli , and Siwei, we have combined ideas from QFT and Condensed Matter to answer the question and checked it extensively in numerics. Surely it has the most beautiful figures in any of my papers.
1
1
42