work

I was definitely in a time crunch this week with pressure mounting on some passing and upcoming deadlines. I got all my S&P reviews finished last week, but there's been some discussion for both TCC and for S&P that I've had to keep my eye on. Fortunately nothing too major, and I think the papers I pushed for are likely to make it all the way. In addition to that work though, I have a big DSO presentation coming up. The report I think went well, but it always takes me a bit to figure out how exactly to present the material to the DSO audience. While I've been working on that, I have my summer undergrad students who are wrapping up next week and a huge office shuffle that happened this week! (I'm now in Gateway South 453 in case someone needs to find me. :-)) I'm also traveling this weekend, so I don't have time on Friday and Monday to do much. I think I've gotten what I need to done; however, I'll definitely need to practice my timing and refine the slides a bit more before the big presentation Tuesday night.

Beyond that, there is the looming Eurocrypt deadline. I'm hoping to prepare a few papers to submit there, but the past month essentially has been filled with travel and other distractions. In about a week, I also start a brand new class that I honestly need some more time to prepare. So I guess we'll just see how it all shakes out :-). One thing that's very fortunate though, is that I can't submit any of my work with Kevin and Pino to Eurocrypt since Pino is the program chair. Otherwise I'd have even more writing and research to get together.

non-work

Most of my time in NJ was working this week to be honest. Last weekend, I had some fun listening to some vinyl and reading Dungeon Crawler Carl, but during the week, I was very focused on hitting the deadlines I have to. That's mostly because this weekend I'm visiting Greenville! I have a family friend who is getting married this weekend, so I decided to turn the trip into a long weekend to see my family, many of whom I haven't seen in months. Fortunately, I worked hard enough that I can (mostly) avoid working this weekend. It's important to me that I can spend time focused on my family and the isolate the work stress just to when I'm back in NJ.

questions

• Has anyone built a batch PIR that is efficient for any batch size?
  • I think asymptotically this isn't too hard, since the hashing/bucketting can be done by the server at query time. But I don't think that'd really be concretely efficient
  • Is there some single preprocessing that someone could do to support batches of any size in a concretely efficient way? (Or maybe avoiding the overhead introduced by online hashing?)
• What do most PhD students think about the recent advances in AI?
  • I've talked with Gabe a bit this week, and it seems like there's a lot of consternation around it
  • I personally feel like I have a very pragmatist view, where even if we no longer prove things ourselves, we might still have an important role in choosing what gets proven
  • Of course, there's always the AGI-argument that any thing a human can do, an AGI in principle could do better, but maybe I'm also just an optimist

Thanks for reading to the end! Here's is Timmy wanted to play while I work this week and a picture of a (very tasty) frozen mocha.