Kickback Cuts Backprop's Red-Tape: Biologically Plausible Credit Assignment in Neural Networks

By David Balduzzi, Hastagiri P. Vanchinathan, Joachim M. Buhmann. 2014

74 citations recorded by the source corpus.

Error backpropagation is an extremely effective algorithm for assigning credit in artificial neural networks. However, weight updates under Backprop depend on lengthy recursive computations and require separate output and error messages — features not shared by biological neurons, that are perhaps unnecessary. In this paper, we revisit Backprop and the credit assignment problem. We first decompose Backprop into a collection of interacting learning algorithms; provide regret bounds on the performance of these sub-algorithms; and factorize Backprop's error signals. Using these results, we derive a new credit assignment algorithm for nonparametric regression, Kickback, that is significantly simpler than Backprop. Finally, we provide a sufficient condition for Kickback to follow error gradients, and show that Kickback matches Backprop's performance on real-world regression benchmarks.

Subfields: Artificial Intelligence, Cognitive Neuroscience, Optimization & Training, Neuroscience

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Cognitive Neuroscience

In Optimization & Training

In Neuroscience

Citations in the corpus

Training Neural Networks Using Features Replay (2018, background): “[1, 27] broke the local dependencies between successive layers and made all hidden layers receive error information from the output layer directly.”
The Shattered Gradients Problem: If resnets are the answer, then what is the question? (2017, background): “It was shown in Balduzzi et al. (2015); Balduzzi (2016) that the gradients in neural nets are sums of path-weights over active paths.”
Decoupled Parallel Backpropagation with Convergence Guarantee (2018, background): “(Nøkland, 2016; Balduzzi et al., 2015) broke the local dependencies between successive layers and made all hidden layers receive error information from the output layer directly.”
Sparse Unsupervised Capsules Generalize Better (2018, background): “Despite a thorough search (Bengio et al., 2016; Balduzzi et al., 2015), researchers have failed to find a biological equivalent to credit assignment via error-gradient backpropagation.”
Greedy InfoMax for Biologically Plausible Self-Supervised Representation Learning (2019, methodology): “There are a number of optimization algorithms that eliminate the need for backpropagation altogether [Balduzzi et al., 2015, Kohan et al., 2018, Lee et al., 2015, Lillicrap et al., 2016, Scellier and Bengio, 2017]. In contrast to our method, these methods employ a global supervised loss function and”
Theories of Error Back-Propagation in the Brain (2019, background): “There is a diversity of ideas on how the back-propagation algorithm may be approximated in the brain [31 – 36]; however, we review the principles behind a set of related models [11,13,14,37] that have substantial connections with biological data while closely paralleling the back-propagation…”
Putting An End to End-to-End: Gradient-Isolated Learning of Representations (2019, methodology): “8] when applied to gradient-isolated modules, freeing the method from end-to-end backpropagation. There are a number of optimization algorithms that eliminate the need for backpropagation altogether [Balduzzi et al., 2015, Kohan et al., 2018, Lee et al., 2015, Lillicrap et al., 2016, Scellier and Be”
Associated Learning: Decomposing End-to-End Backpropagation Based on Autoencoders and Target Propagation (2019, methodology): “e can be updated independently. increments based on a single objective [9]. Additionally, some components that are essential for backpropagation to work correctly have not been observed in the cortex [10]. Therefore, many works have proposed methods that more closely resemble the operations of biolo”
An Approximation of the Error Backpropagation Algorithm in a Predictive Coding Network with Local Hebbian Synaptic Plasticity (2017, background): “Balduzzi et al. (2014) showed that efficient learning may be achieved by a network that receives a global error signal and in which synaptic weight modification depends jointly on the error and the terms describing the influence of each neuron of final error.”
Material for Decoupled Neural Interfaces using Synthetic Gradients (2017, methodology): “, 2016), Direct Feedback Alignment (Nøkland, 2016), and Kickback (Balduzzi et al., 2014)) – interestingly Czarnecki et al.”
Understanding Synthetic Gradients and Decoupled Neural Interfaces (2017, methodology): “, 2016), and Kickback (Balduzzi et al., 2014), and show that all these error approximation schemes can be captured in a unified framework, but crucially only using synthetic gradients can one achieve unlocked training.”