Biological credit assignment through dynamic inversion of feedforward networks
arXiv:2007.05112
Abstract
Learning depends on changes in synaptic connections deep inside the brain. In multilayer networks, these changes are triggered by error signals fed back from the output, generally through a stepwise inversion of the feedforward processing steps. The gold standard for this process -- backpropagation -- works well in artificial neural networks, but is biologically implausible. Several recent proposals have emerged to address this problem, but many of these biologically-plausible schemes are based on learning an independent set of feedback connections. This complicates the assignment of errors to each synapse by making it dependent upon a second learning problem, and by fitting inversions rather than guaranteeing them. Here, we show that feedforward network transformations can be effectively inverted through dynamics. We derive this dynamic inversion from the perspective of feedback control, where the forward transformation is reused and dynamically interacts with fixed or random feedback to propagate error signals during the backward pass. Importantly, this scheme does not rely upon a second learning problem for feedback because accurate inversion is guaranteed through the network dynamics. We map these dynamics onto generic feedforward networks, and show that the resulting algorithm performs well on several supervised and unsupervised datasets. Finally, we discuss potential links between dynamic inversion and second-order optimization. Overall, our work introduces an alternative perspective on credit assignment in the brain, and proposes a special role for temporal dynamics and feedback control during learning.
arXiv comment34th Conference on Neural Information Processing Systems (NeurIPS 2020), Vancouver, Canada
References in corpus (15)
- Exact solutions to the nonlinear dynamics of learning in deep linear neural networks
- Identifying and attacking the saddle point problem in high-dimensional non-convex optimization
- New insights and perspectives on the natural gradient method
- Direct Feedback Alignment Provides Learning in Deep Neural Networks
- How Auto-Encoders Could Provide Credit Assignment in Deep Networks via Target Propagation
- Assessing the Scalability of Biologically-Motivated Deep Learning Algorithms and Architectures
- Deep Learning without Weight Transport
- Natural Neural Networks
- Feedback alignment in deep convolutional networks
- Biologically-plausible learning algorithms can scale to large datasets
- Practical Gauss-Newton Optimisation for Deep Learning
- A Theoretical Framework for Target Propagation
- Learning to solve the credit assignment problem
- Deriving Differential Target Propagation from Iterating Approximate Inverses
- Error Forward-Propagation: Reusing Feedforward Connections to Propagate Errors in Deep Learning