3 ms·

I think I'm an even older fart, and a lot of this resonates with me. So a few thoughts. First, I've gone back to school. I have a PhD in computer science from

I think I'm an even older fart, and a lot of this resonates with me. So a few thoughts. First, I've gone back to school. I have a PhD in computer science from 1983, in which I explored spatial query processing. (General purpose ideas, but mostly applied to database systems.) My PhD has basically expired, the amount of new stuff to learn is daunting, and for a variety of other reasons, I thought it would be better to go back to school rather than study on my own. So I'm enrolled in a CS MS program to learn about AI and cognitive psychology. Second: To a first approximation, I view the systemsy parts of computer science: CPUs, compilers, operating systems, data structures and algorithms, as having served their purpose, which was to enable AI in its current form. AI is just a different discipline. I happen to be familiar with some parts of those foundations, but I'm basically starting over with AI. Finally: I recall what I now realize was a fork in the road. I took an AI course in 1980, I think, and we learned about A*, and simple feature detection in images, and alpha-beta pruning, and so on. Symbolic AI was up and coming, and looked to be the way to go. I remember learning about perceptrons and being intrigued. And then I learned about Minsky's proof that a perceptron can't compute XOR, so this idea of computing with a neurons was basically a dead end. Also, I liked the precision of algorithms: the answer is right or not. It has such and such worst case running time. I disliked the wishy washiness of AI, with uncertain and only probably correct answers. I distrusted the idea of tall stacks of probability calculations yielding anything other than a guess. And so I picked my path (into data structures, algorithms, databases). Now, of course, I realize how wrong it was to dismiss the path of neural models of computation, and the usefulness of statistical methods. I have no regrets, I've had a very satisfying career. But if I were starting out now, I would study a lot more math, and hell yes, get deep into machine learning.

erikerikson 3y ago
An update on Minsky's proof. It was correctly formulated but based on a too limited model. Adding location into the model allows learning XOR. I demonstrated this with Hebbian learning on a Hopfield network finding learning of up to 6 dimensional XORs. This was a replication of work by Elman and reported in Rethinking Innateness[0]. A more provocative claim: layers and back propagation, despite providing us with fantastic advancements, are unnecessary. [0] https://mitpress.mit.edu/9780262050524/rethinking-innateness/ https://mitpress.mit.edu/9780262050524/rethinking-innateness...
alex_suzuki 3y ago
You‘re definitely right on one count: you are an even older fart than me. ;-) Just kidding, but I’ve actively encouraged my 8yr old daughter to learn how to code (using Scratch) and beginning to think that might be a dead-end… hope she will enjoy math!
- geophile 3y ago
  My daughter graduated with a degree in physics and astronomy, and then went to work for Epic (the medical record systems company). She has experience in Python from school, ancient stuff from Epic (related to MUMPS), and has been teaching herself C# and other technologies. She is thinking about more computer science education, on her own, with the aim of becoming a software engineer. I'm not at all sure that's a wise decision, right now, especially with the glut of software engineers, with far more experience, being laid off by FAANG. She's leaving EPIC, and taking a several months long road trip with her boyfriend, and will find a job when she comes off the road. She's a smart kid, she'll figure it out. I just don't think she will end up in software.
  - alex_suzuki 3y ago
    She sounds awesome! Thank you for taking the time to compose such thoughtful replies.
  - insane_dreamer 3y ago
    sounds similar to my daughter's path, except she did chem-eng instead of physics; joined Epic out of college now going on 5 years and in that time has switched over to full-time SW engineering (mostly SQL, Typescript). She enjoys that but I'm less sure of its value (unless she becomes an Epic lifer or someone maintaining Epic systems at hospitals, which is what a lot of former employees end up doing and they make more $ that way), compared to continuing on with hard sciences.
    - geophile 3y ago
      Working at EPIC, you get familiar with an extremely important yet obscure part of the software world: HL7, MUMPS. I suspect that my daughter will be able to captialize on that knowledge. She does not want to stay at EPIC. Is yoiur duaghter happy there?

erikerikson 3y ago

An update on Minsky's proof. It was correctly formulated but based on a too limited model. Adding location into the model allows learning XOR. I demonstrated this with Hebbian learning on a Hopfield network finding learning of up to 6 dimensional XORs. This was a replication of work by Elman and reported in Rethinking Innateness[0]. A more provocative claim: layers and back propagation, despite providing us with fantastic advancements, are unnecessary. [0] https://mitpress.mit.edu/9780262050524/rethinking-innateness/ https://mitpress.mit.edu/9780262050524/rethinking-innateness...

alex_suzuki 3y ago

You‘re definitely right on one count: you are an even older fart than me. ;-) Just kidding, but I’ve actively encouraged my 8yr old daughter to learn how to code (using Scratch) and beginning to think that might be a dead-end… hope she will enjoy math!

geophile 3y ago
My daughter graduated with a degree in physics and astronomy, and then went to work for Epic (the medical record systems company). She has experience in Python from school, ancient stuff from Epic (related to MUMPS), and has been teaching herself C# and other technologies. She is thinking about more computer science education, on her own, with the aim of becoming a software engineer. I'm not at all sure that's a wise decision, right now, especially with the glut of software engineers, with far more experience, being laid off by FAANG. She's leaving EPIC, and taking a several months long road trip with her boyfriend, and will find a job when she comes off the road. She's a smart kid, she'll figure it out. I just don't think she will end up in software.
- alex_suzuki 3y ago
  She sounds awesome! Thank you for taking the time to compose such thoughtful replies.
- insane_dreamer 3y ago
  sounds similar to my daughter's path, except she did chem-eng instead of physics; joined Epic out of college now going on 5 years and in that time has switched over to full-time SW engineering (mostly SQL, Typescript). She enjoys that but I'm less sure of its value (unless she becomes an Epic lifer or someone maintaining Epic systems at hospitals, which is what a lot of former employees end up doing and they make more $ that way), compared to continuing on with hard sciences.
  - geophile 3y ago
    Working at EPIC, you get familiar with an extremely important yet obscure part of the software world: HL7, MUMPS. I suspect that my daughter will be able to captialize on that knowledge. She does not want to stay at EPIC. Is yoiur duaghter happy there?