Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
datameta
on Oct 8, 2023
|
parent
|
context
|
favorite
| on:
Tiny Language Models Come of Age
Is it possible to use ML to improve on training returns? Similar to neural architecture search or circuit design optimization, can we use it to maximize performance and accuracy through training set selection/optimization?
woadwarrior01
on Oct 8, 2023
[–]
Data subset selection is an age old problem / sub-field in ML. Feature selection is a special case of this broader problem.
Join us for
AI Startup School
this June 16-17 in San Francisco!
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: