It depends on the network, @deworrall92 and others have done nice stuff on rotation equivariance e.g. https://deworrall92.github.io/2017/06/01/harmonic_networks.html … :)
-
-
-
still, it looks like its inference roughly corresponds with mine (except when the 'duck' is upside down - need to do a better job with those priors). We certainly aren't rotationally equivariant!
- 1 more reply
New conversation -
-
- 1 more reply
New conversation -
-
It’s been such an intense kind of sociology week
@LivUniSoc that here’s Wittgenstein turning up in my Twitter feed first thing on a Saturday morning? Think I’m still dreaming. -
The only two words I remember from those who did sociology at
@GoldsmithsUoL were phenomenology and ethnomethodology. - 3 more replies
New conversation -
-
-
The answer is simple: Yes!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
But what about when the duckrabbits are going at it like bunnies?pic.twitter.com/nAS3243CFw
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Dammit, so which one is it? :)
-
We can't easily say it's both because they're incompatible with each other. But to then say that it is neither a duck nor a rabbit misses something, as either interpretation of the black &white dots is an adequate interpretation. Aristotle's logic replaced by a logic of ambiguity
- 1 more reply
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.