# GAIT-prop: A biologically plausible learning rule derived from backpropagation of error

arXiv: https://arxiv.org/abs/2006.06438 · HTML: https://mlsanity.com/paper/2006.06438/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Jun 2020.

## Built on by (most cited first, 15 of 15)

- [Holomorphic Equilibrium Propagation Computes Exact Gradients Through Finite Size Oscillations](https://mlsanity.com/paper/2209.00530/) (2022)
- [Blockwise Self-Supervised Learning at Scale](https://mlsanity.com/paper/2302.01647/) (2023)
- [Single-phase deep learning in cortico-cortical networks](https://mlsanity.com/paper/2206.11769/) (2022)
- [Predictive coding feedback results in perceived illusory contours in a recurrent neural network](https://mlsanity.com/paper/2102.01955/) (2021)
- [Meta-Learning Bidirectional Update Rules](https://mlsanity.com/paper/2104.04657/) (2021)
- [Constrained Parameter Inference as a Principle for Learning](https://mlsanity.com/paper/2203.13203/) (2022)
- [Why Linguistics Will Thrive in the 21st Century: A Reply to Piantadosi (2023)](https://mlsanity.com/paper/2308.03228/) (2023)
- [Fixed-Weight Difference Target Propagation](https://mlsanity.com/paper/2212.10352/) (2022)
- [Gradient-Free Training of Recurrent Neural Networks using Random Perturbations](https://mlsanity.com/paper/2405.08967/) (2024)
- [Cortico-cerebellar networks as decoupling neural interfaces](https://mlsanity.com/paper/2110.11501/) (2021)
- [Target Propagation via Regularized Inversion](https://mlsanity.com/paper/2112.01453/) (2021)
- [Gradient-adjusted Incremental Target Propagation Provides Effective Credit Assignment in Deep Neural Networks](https://mlsanity.com/paper/2102.11598/) (2021)
- [Information Bottleneck-Based Hebbian Learning Rule Naturally Ties Working Memory and Synaptic Updates](https://mlsanity.com/paper/2111.13187/) (2021)
- [Biologically-Motivated Learning Model for Instructed Visual Processing](https://mlsanity.com/paper/2306.02415/) (2023)
- [A Unified Perspective on Optimization in Machine Learning and Neuroscience: From Gradient Descent to Neural Adaptation](https://mlsanity.com/paper/2510.18812/) (2025)

## Abstract (arXiv)

Traditional backpropagation of error, though a highly successful algorithm for learning in artificial neural network models, includes features which are biologically implausible for learning in real neural circuits. An alternative called target propagation proposes to solve this implausibility by using a top-down model of neural activity to convert an error at the output of a neural network into layer-wise and plausible 'targets' for every unit. These targets can then be used to produce weight updates for network training. However, thus far, target propagation has been heuristically proposed without demonstrable equivalence to backpropagation. Here, we derive an exact correspondence between backpropagation and a modified form of target propagation (GAIT-prop) where the target is a small perturbation of the forward pass. Specifically, backpropagation and GAIT-prop give identical updates when synaptic weight matrices are orthogonal. In a series of simple computer vision experiments, we show near-identical performance between backpropagation and GAIT-prop with a soft orthogonality-inducing regularizer.