It is reminiscent of work I did on backprop-free DL for CV (in 2010, 2012 and 2015 ), but I could never get my algo to scale to many layers
Unsure what to think about Stacked Regression Machines http://arxiv.org/abs/1608.04062 Did anyone repro, try on more datasets?
-
-
-
I had checked out the PCANet results when they got out but it never seemed very convincing to me at the time
End of conversation
New conversation -
-
-
ask
@HonghuiS (one of the authors) for the code -
plans to put it on Github?
End of conversation
New conversation -
-
-
Smells a bit fishy to me too. arXiv submission only after NIPS acceptance feels like an end run around reproduction efforts.
-
I'd love to see a reusable OS implementation. I want to believe
- Show replies
New conversation -
-
-
this can be a pretty big deal if what they claim can be reproduced.
-
for sure. I'd be excited if that were the case. Especially since it would show how I could make my algo work at scale
End of conversation
New conversation -
-
-
Just noticed: "arXiv admin note: text overlap with arXiv:1404.3606 by other authors" Doesn't bode so well...
-
it'd be useful if they highlighted the text that overlaps
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.