Transfer entropy-based feedback improves performance in artificial neural networks
This addresses the challenge of improving efficiency and brain-like architecture in artificial neural networks, though it appears incremental as it builds on existing concepts of feedback and transfer entropy.
The paper tackled the problem of achieving top-level performance on a standard benchmark task with small neural networks by incorporating feedback connections, showing that a few-layer network with feedback can match the performance of large feed-forward structures.
The structure of the majority of modern deep neural networks is characterized by uni- directional feed-forward connectivity across a very large number of layers. By contrast, the architecture of the cortex of vertebrates contains fewer hierarchical levels but many recurrent and feedback connections. Here we show that a small, few-layer artificial neural network that employs feedback will reach top level performance on a standard benchmark task, otherwise only obtained by large feed-forward structures. To achieve this we use feed-forward transfer entropy between neurons to structure feedback connectivity. Transfer entropy can here intuitively be understood as a measure for the relevance of certain pathways in the network, which are then amplified by feedback. Feedback may therefore be key for high network performance in small brain-like architectures.