In their new book "If Anyone Builds It, Everyone Dies," Yudkowsky and Soares argue that we can view an Artificially Super Intelligent agent as though it has its own agency and desires. But computers are a different beast.
Because of zero-knowledge SNARKs that rely on our sumcheck protocol, Algebraic Methods for Interactive Proof Systems (1992) with Lund, Karloff and Nisan is now my most cited paper edging out NEXP = MIP (1991) with Babai and Lund.
Whenever someone tells me "wait until you see what quantum and AI will do together", I'm reminded of this passage from the book Harry Potter and the Deathly Hallows.
⏰ The 🧑🎓 travel support application and 👶 childcare support applications are both due in ~25h!
The former is open to all students and postdocs (not necessarily authors of a paper). The latter provides financial support (to be used in any childcare-related way) to conference attendees!
When I went to college in the early 80's, it was a faux pas to wear a backpack using both straps. Four decades later I still find myself slinging my backpack over one shoulder.
Took my own advice and rewatched Sneakers. This line aged well.
There's a war out there, old friend. A world war. And it's not about who's got the most bullets. It's about who controls the information. What we see and hear, how we work, what we think. It's all about the information!
Q: Suppose we have a Turing machine with a stay-put option, δ:QxΓ→QxΓx{L,R,S}. Can you create an equivalent Turing machine with only L,R moves without enlarging Q or Γ ?
Want to remember Robert Redford - Sneakers, the 1992 movie where his character led a professional hacking team going after a device that will break any code. Better than it sounds.
I gave a homework bonus question that even AI couldn't solve. Can you?
Suppose we have a Turing machine with a stay-put option, δ:QxΓ→QxΓx{L,R,S}. Can you create an equivalent Turing machine with only L,R moves without enlarging Q or Γ ?
