Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
bob1029
50 days ago
|
parent
|
context
|
favorite
| on:
AMD acquires Taalas to boost inference performance...
I feel like NAND process tech could become useful at solving some of these problems. A GPU where you can update the weights a few thousand times may be sufficient.
mdp2021
50 days ago
|
next
[–]
The basis of Taalas is "compute in memory" electronics - past Von Neumann's separation of processor and memory.
You need to be able to add|mul where the data (the weights) are stored.
addaon
50 days ago
|
prev
|
next
[–]
NAND hasn't been scaling great lately. It seems like PCM or MRAM would both be better fits.
kridsdale1
50 days ago
|
prev
[–]
FPGA model storage?
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: