those circuits are trainable?
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Yea and it would be interesting to see how well the optimization techniques would work here. Simulated annealing comes to mind.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Could you elaborate?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
You start on the ground and work your way up, layer after layer, adding some tasty garlic and other spices here and there
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
you may find this video about designing with LMS (pre-SGD approach to finding weights) interesting https://youtu.be/skfNlwEbqck
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
"Analog VLSI and Neural Systems" by Carver Mead talked about this, but the field hasn't caught up to theoretical deep learning yet
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This post reminded me of a comment from a speech researcher about
#WaveNet . -
WaveNet is analogous to the change of signal processing from analogue to digital. Analog = IIR filter, digital = FIR filter.
- Show replies
New conversation -
-
-
Yep! Carver Mead thought ahead of his time, one might say (the beginning of neuromorphic chips, which failed then).
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This is my research topic. Exciting news to come!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.