no code implementations • 13 Jan 2024 • Joseph Pemberton, Rui Ponte Costa
Inspired by the accumulate $\mathrm{TD}(\lambda)$ in RL, we propose a fully online method for learning synthetic gradients which avoids the use of BPTT altogether: accumulate $BP(\lambda)$.
1 code implementation • 23 Jun 2022 • Will Greedy, Heng Wei Zhu, Joseph Pemberton, Jack Mellor, Rui Ponte Costa
First, we demonstrate that our model can effectively backpropagate errors through multiple layers using a single-phase learning process.
1 code implementation • NeurIPS 2021 • Joseph Pemberton, Ellen Boven, Richard Apps, Rui Ponte Costa
Recently, decoupled neural interfaces (DNIs) were introduced as a solution to the forward and feedback locking problems in deep networks.
no code implementations • 1 Jan 2021 • Joseph Pemberton, Ellen Boven, Richard Apps, Rui Ponte Costa
Deep learning methods suffer from similar locking constraints both on the forward and backward phase.