Keller Jordan @kellerjordan0 · 13 Apr 2023
3/ However, it’s not yet clear if these lucky runs of training are actually better than unlucky ones. To find out, I split the CIFAR-10 test-set into two halves, and evaluated thousands of trained networks against both. Intuitively, lucky networks should do well on both splits.
7 420Views
44Likes
0Reposts
1Replies
0Quotes
0Bookmarks
Is that a lot?
0.79×vs this author's median9 348 views is typical
0Percentile for this authorof 5 recent posts
5.10×vs 10K–100K median1 456 views is typical
38.41%Reachviews ÷ followers
0.61%Engagement rateof viewers reacted