diff --git a/_posts/2023-10-26-Thursday-October-26th.md b/_posts/2023-10-26-Thursday-October-26th.md new file mode 100644 index 0000000000000..3808c6d6f70fc --- /dev/null +++ b/_posts/2023-10-26-Thursday-October-26th.md @@ -0,0 +1,104 @@ +## My first SQuInT and trip to NM + +## I'm currently at SQuInT, but what is it? +- It's a SouthWest Symposium in Quantum, and so far it's been awesome! (day 1) +- I got in really late last night after three flights, and the computer system in the hotel had gone down, +so they had to manually check me in, but they were so nice, and super efficient, and the check-in process +was fantastic. People were also really nice; my ride from the airport was also really nice; I asked the driver +about places to eat, places to go, and the general economy of the area, and it was just really great. + +## Turn around +- So I got to bed around 1am, but realized that the symposium was also in the same hotel where I was staying, +which meant I could sleep in until 7 and go down for 7:30am breakfast. Once I went down to the eating area, +I realized I knew two people from other workshops / conferences, including someone I was working with on +a current project, who lives in California and was presenting a poster. This is my first year, and I got +a fellowship to attend, so I'm super thrilled. My goal for next year would be to present a poster, which +I think given the scope of what I'm currently working on, I could probably do (the deadline this year +was in August, so by the time I found out about the workshop during summer school, it was too close +to the deadline!). That being said, I'm also submitting to some other things at the moment, and it's +been a super busy semester for me in general, juggling so many projects, but also very rewarding and +I think that this (what I'm doing now) is what I want to stick with in terms of the community and what I like doing. +- The community is intimate, and full of theorists, and people are nice, and it's really felt like a great fit +for me and the projects I've been working on. +- In the evening was the poster session. So definitely going to visit my collaborator's poster! She also +gave me a really great resource for a current problem I'm tackling, too! + +## Notes +- The keynote was from Bill. The question: Have we seen a demonstration of experimental quantum advantage? +We want to find problems that can be solved using a near-term quantum experiment that are classically hard +to solve i.e. they can't be solved in polynomial time. And we want our solution to be efficiently verified +with a classical computer with minimal trust in the experiment. Quantum has focused on "sampling problems" +in which the goal is to sample from a complicated distribution. Current experiments are not scalable +and the verification is not efficient and noise causes the signal to rapidly decay. +- The Goldilocks parameter: these issues force the current quantum advantage candidates to find this. +Stockmeyer reduction and P-sharp hard problems. Lipton ('91) shows the average case hardness of permanent +vs random nature. Questions: How close are the output distribution of noisy (i.e. depolarizing) random curcuit +and uniform distribution? +- The anti-concentration property: means that we are not concentrated in one area, is sufficiently spread. +It is a property of sufficiently deep random quantum circuits. +- Open: In recent work (Ghosh et al, 23), we show that circuit distributions never anti-concentrate for +random circuits with hybrid noise. +- Measure and Forget Dynamics in Random Circuits: we measure, but "Forget" the outcome. We "dephase" and +"forget" the matrix. Application on Yoshida's Decoding Protocol. See "How Dynamical Quantum Memories Forget" +by Fidkowski et al, and "Decoding the Entanglement Structure of Monitored Quantum Circuits" by Yoshida. +This method focuses on Clifford, not Haar circuits. +- Shivan's talk on "Arbitrary Random Quantum Circuits from Unitary Designs". Unitary designs are modelled +as random walk. How close is this distribution to normal uniform distribution? One step random walk related +to the spectral gap of the 2-local, frustration-free Hamiltonian. Generalizing Knabe method for Spectral +gap. First approach uses a star-graph approach versus the spanning-tree graph (Dectectability Lemma) and the +Quantum Union Bound (we relate this to the spanning tree to a 1D graph). I really liked this talk and will +check out the paper! +- Tobias's talk on "A Wannier-Stark Optical Lattice Clock with Extended Coherence times". Goal of precision +in say, meteorology. We distinguish between accuracy and precision, where accuracy is defined as relating +to how centred we are on a target and precision thinks about the spread. Intro to atomic clocks using +Strontium, where ```v_0``` is our oscillator frequency. (Tobias also showed another possible Yterbium +clock, I believe by NIST). We also saw an example of perturbation on our clock measurement on the Bloch +sphere. +- Hayden gave a talk on "Portable Atomic Clocks" (from Sandia). Question: What is the tradeoff in +size/ weight / power for something liek a satellite or vehicle atomic clock. Second project after +the primary TicToc one on Robust Optical Clock Networks. +- Yu Chen from Google gave a talk on "Beyond classical quantum computing". I had a meeting with my +abelian varieties problem-set group and then a panel on navigating grad school with some mathematicians, +so I had to skip a couple talks and then return for the poster session today. + + +## Here are some photos from the first day. + + + + + + + + + + + + + + + + + + + + + + + + + +- Someone was making paper cranes from the black tissues apparently. It's been a very nice +group of people. We got into talking about Oppenheimer and "American Prometheus" and the +IAS scene in the movie, as well as the 1937 paper that was mentioned in the movie, but that +didn't really add to the movie, which was hilarious. People had strong thoughts! + + + +- Some very good flan! We also had tostadas and some other good food for lunch! + + + + + +## And that's it