The many directions of feedback alignment

Cheung, Brian; Jiang, Daniel L.

doi:10.32470/ccn.2018.1241-0

Cited by 1 publication

(1 citation statement)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Alignment schemes such as weight mirrors have yielded promising results on large extensive benchmarks, all without requiring weight transport and even when incorporated into convolutional frameworks. Very recent credit assignment schemes have combined the notion of weight mirroring in the context of complementary (global) forward and backward pathways [62,436] (this scheme combines RFA with concepts of target propagation [30], reviewed later). Other variant schemes take the sign of the forward activities instead, such as the 'sign symmetry' method [462], where the feedback synapses that generate the requisite teaching signals share only the sign but not the magnitude of the feedforward ones [243].…”

Section: Global Explicit Signal Algorithmsmentioning

confidence: 99%

Brain-Inspired Machine Intelligence: A Survey of Neurobiologically-Plausible Credit Assignment

Ororbia

2023

Preprint

View full text Add to dashboard Cite

In this survey, we examine algorithms for conducting credit assignment in artificial neural networks that are inspired or motivated by neurobiology, unifying these various processes under one possible taxonomy. Our proposed taxonomy is constructed based on how a learning algorithm answers a central question underpinning the mechanisms of synaptic plasticity in complex adaptive neuronal systems: where do the signals that drive the learning in individual elements of a network come from and how are they produced? In this unified treatment, we organize the ever-growing set of brain-inspired learning processes into six general families and consider these in the context of backpropagation of errors and its known criticisms. The results of this review are meant to encourage future developments in neuro-mimetic systems and their constituent learning processes, wherein lies an important opportunity to build a strong bridge between machine learning, computational neuroscience, and cognitive science.

show abstract

Section: Global Explicit Signal Algorithmsmentioning

confidence: 99%

Brain-Inspired Machine Intelligence: A Survey of Neurobiologically-Plausible Credit Assignment

Ororbia

2023

Preprint

View full text Add to dashboard Cite

show abstract

The many directions of feedback alignment

Cited by 1 publication

References 5 publications

Brain-Inspired Machine Intelligence: A Survey of Neurobiologically-Plausible Credit Assignment

Brain-Inspired Machine Intelligence: A Survey of Neurobiologically-Plausible Credit Assignment

Contact Info

Product

Resources

About