It looks like there are still reverse signals (eg., deltas), but they are multiplied by a random matrix B instead of the usual transpose of the weight matrix for that layer. Am I misunderstanding?
I believe instead of multiplying the delta by W^t to backpropagate the error from layer l to l-1, you multiply it by a random projection B. It's hard to dig deeper because there doesn't appear to be any other…
It looks like there are still reverse signals (eg., deltas), but they are multiplied by a random matrix B instead of the usual transpose of the weight matrix for that layer. Am I misunderstanding?
I believe instead of multiplying the delta by W^t to backpropagate the error from layer l to l-1, you multiply it by a random projection B. It's hard to dig deeper because there doesn't appear to be any other…