HN user

Grambo

42 karma
Posts0
Comments8
View on HN
No posts found.

I loved the Southern Reach trilogy but didn't finish TINAD. I thought the premise was pushed too far and that less would've been more. OTOH the atmosphere of oppressive bureaucracy of Authority is still one my all time favorite scifi reads.

I experienced something like this recently. I struggled to grasp linear algebra when I took it in undergrad and as a result was always intimidated by the subject. Now, years later, I'm taking a course in graphics and naturally needing to relearn linear algebra and suddenly everything just clicks.

My understanding of the use of FPGAs and ASICs that are used to speed up neural networks (such as those in phones) is that they are simply designed to do the types of calculations used for NNs more quickly (matrix operations) and generally at a reduced level of precision. This is very different from a memristor approach where the structure of the network itself would be represented in the silicon. I also think it's unfair to compare the two because it took decades of work to get CMOS transistors to where they are today. I imagine that once commercial applications for memristors appear many optimizations/improvements will present themselves.

Actually I think there's a reason for this. I took a cognitive science class once where our professor said that many believe that the portion of the brain responsible for recognizing faces is separate from the portion that does all other visual recognition. Perhaps that's why people have a hard time remembering what a face looks like. It's also why some people can lose their ability to recognize faces but can still recognize all other objects.