Learning representations for binary-classification without backpropagation
Lechner M. 2020. Learning representations for binary-classification without backpropagation. 8th International Conference on Learning Representations. ICLR: International Conference on Learning Representations.
Download
Download (ext.)
https://openreview.net/forum?id=Bke61krFvS
[Published Version]
Conference Paper
| Published
| English
Scopus indexed
Author
Corresponding author has ISTA affiliation
Department
Grant
Abstract
The family of feedback alignment (FA) algorithms aims to provide a more biologically motivated alternative to backpropagation (BP), by substituting the computations that are unrealistic to be implemented in physical brains. While FA algorithms have been shown to work well in practice, there is a lack of rigorous theory proofing their learning capabilities. Here we introduce the first feedback alignment algorithm with provable learning guarantees. In contrast to existing work, we do not require any assumption about the size or depth of the network except that it has a single output neuron, i.e., such as for binary classification tasks. We show that our FA algorithm can deliver its theoretical promises in practice, surpassing the learning performance of existing FA methods and matching backpropagation in binary classification tasks. Finally, we demonstrate the limits of our FA variant when the number of output neurons grows beyond a certain quantity.
Publishing Year
Date Published
2020-03-11
Proceedings Title
8th International Conference on Learning Representations
Publisher
ICLR
Acknowledgement
This research was supported in part by the Austrian Science Fund (FWF) under grant Z211-N23
(Wittgenstein Award).
Conference
ICLR: International Conference on Learning Representations
Conference Location
Virtual ; Addis Ababa, Ethiopia
Conference Date
2020-04-26 – 2020-05-01
IST-REx-ID
Cite this
Lechner M. Learning representations for binary-classification without backpropagation. In: 8th International Conference on Learning Representations. ICLR; 2020.
Lechner, M. (2020). Learning representations for binary-classification without backpropagation. In 8th International Conference on Learning Representations. Virtual ; Addis Ababa, Ethiopia: ICLR.
Lechner, Mathias. “Learning Representations for Binary-Classification without Backpropagation.” In 8th International Conference on Learning Representations. ICLR, 2020.
M. Lechner, “Learning representations for binary-classification without backpropagation,” in 8th International Conference on Learning Representations, Virtual ; Addis Ababa, Ethiopia, 2020.
Lechner M. 2020. Learning representations for binary-classification without backpropagation. 8th International Conference on Learning Representations. ICLR: International Conference on Learning Representations.
Lechner, Mathias. “Learning Representations for Binary-Classification without Backpropagation.” 8th International Conference on Learning Representations, ICLR, 2020.
All files available under the following license(s):
Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported (CC BY-NC-ND 3.0):
Main File(s)
File Name
iclr_2020.pdf
249.43 KB
Access Level
Open Access
Date Uploaded
2022-01-26
MD5 Checksum
ea13d42dd4541ddb239b6a75821fd6c9
Link(s) to Main File(s)
Access Level
Open Access