![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
guslOn Thursday, Jan 30, I saw Feynman's friends Ralph Leighton, Ed Fredkin and (gasp!) Marvin Minsky at play about Feynman. They were very casual. I stuck around and chatted with Minsky and his student Push Singh. I ended up having dinner with them. Push said I should stop by the Media Lab, which I did twice the next week. He introduced me to Hugo Liu, another PhD student who needs somebody to help finding and translating a piece of code for the OpenMind Project (Minsky's Common Sense software that uses semantic networks (his answer to the AI problem)), and I offered to help. I'll probably start soon. Just before leaving to Cornell, I attended to the first lecture of Minsky's Common Sense course. He's not the most articulate speaker around, but he's entertaining by being scathing:
(paraphrasing)
"Chomsky is a very productive guy: for 30 years, he single-handedly managed to stop all progress in the science of semantics."
"The problem of the world is religion, and similar irrationality."
"The concept of 'emergence' explains nothing. Those who use it have given up understanding the world."
"The world of AI is composed of 20 000 people wasting time and 20 people making real progress. In the past the popular wrong path was Neural Networks. These days it's Bayesian Networks."
"Filling in numbers is what you should do at the end of your research, when you've given up trying to understanding things better. A lot of people have that backwards."
Ed Fredkin is a very interesting guy. A physics prof at BU, he never went to college. He spent a year at Caltech at Feynman's invitation. He learned directly from books and from Feynman. He has some unusual ideas about information theory and the universe. He said a lot of physicists think he's crazy. He hosts http://www.digitalphilosophy.org/, where he exposes his philosophy that nature is finite in information.
Earlier that day I had spoken to Manuela Veloso, the CMU prof who builds soccer-playing robots and Dan Dennett, the greatest philosopher alive, who attended her talk. She didn't know who he was, so I told her after he left.
Before that, I attended a talk with about communication in games by philosopher Robert Stalnaker.
After the Media Lab, I went to Cornell, and met with a couple of people from the NuPRL Project (the biggest theorem-prover/verification engine around). I also met with Bart Selman (AI, complexity theory bigshot), who I fortunately managed to impress with good questions and original suggestions. The meeting with Halpern didn't go as well as I had expected. The conversation just wasn't flowing so he said we could communicate by email, and I agreed. I also met with superstar Hopcroft, who decided to stop talking to me after asking me my GPA.
Now I'm back in Brazil, indefinitely. In epistemic first-order logic, I could write
K(Ed(US(d))) /\ ~Ed(K(US(d)))
, meaning "I know that there exists a day in which I will go back to the US, but there does not exist a day in which I know I will go back to the US".
Since I'm not interested in making Brazilian money, I'm trying to focus on my personal growth. I'm trying to study logic with a professor here. She is a proof theorist, and seemed interested in tutoring me, but we only met once. I'm also looking for interesting research projects, and the faculty at UFPE seemed quite open, but nothing has developed yet (then again, I just got back). I still have almost two months left to apply for UvAmsterdam's ILLC Master of Logic program.
(paraphrasing)
"Chomsky is a very productive guy: for 30 years, he single-handedly managed to stop all progress in the science of semantics."
"The problem of the world is religion, and similar irrationality."
"The concept of 'emergence' explains nothing. Those who use it have given up understanding the world."
"The world of AI is composed of 20 000 people wasting time and 20 people making real progress. In the past the popular wrong path was Neural Networks. These days it's Bayesian Networks."
"Filling in numbers is what you should do at the end of your research, when you've given up trying to understanding things better. A lot of people have that backwards."
Ed Fredkin is a very interesting guy. A physics prof at BU, he never went to college. He spent a year at Caltech at Feynman's invitation. He learned directly from books and from Feynman. He has some unusual ideas about information theory and the universe. He said a lot of physicists think he's crazy. He hosts http://www.digitalphilosophy.org/, where he exposes his philosophy that nature is finite in information.
Earlier that day I had spoken to Manuela Veloso, the CMU prof who builds soccer-playing robots and Dan Dennett, the greatest philosopher alive, who attended her talk. She didn't know who he was, so I told her after he left.
Before that, I attended a talk with about communication in games by philosopher Robert Stalnaker.
After the Media Lab, I went to Cornell, and met with a couple of people from the NuPRL Project (the biggest theorem-prover/verification engine around). I also met with Bart Selman (AI, complexity theory bigshot), who I fortunately managed to impress with good questions and original suggestions. The meeting with Halpern didn't go as well as I had expected. The conversation just wasn't flowing so he said we could communicate by email, and I agreed. I also met with superstar Hopcroft, who decided to stop talking to me after asking me my GPA.
Now I'm back in Brazil, indefinitely. In epistemic first-order logic, I could write
K(Ed(US(d))) /\ ~Ed(K(US(d)))
, meaning "I know that there exists a day in which I will go back to the US, but there does not exist a day in which I know I will go back to the US".
Since I'm not interested in making Brazilian money, I'm trying to focus on my personal growth. I'm trying to study logic with a professor here. She is a proof theorist, and seemed interested in tutoring me, but we only met once. I'm also looking for interesting research projects, and the faculty at UFPE seemed quite open, but nothing has developed yet (then again, I just got back). I still have almost two months left to apply for UvAmsterdam's ILLC Master of Logic program.