Hybrid predictive coding: Inferring, fast and slow

Predictive coding is an influential model of cortical neural activity. It proposes that perceptual beliefs are furnished by sequentially minimising “prediction errors”—the differences between predicted and observed data. Implicit in this proposal is the idea that successful perception requires multiple cycles of neural activity. This is at odds with evidence that several aspects of visual perception—including complex forms of object recognition—arise from an initial “feedforward sweep” that occurs on fast timescales which preclude substantial recurrent activity. Here, we propose that the feedforward sweep can be understood as performing amortized inference (applying a learned function that maps directly from data to beliefs) and recurrent processing can be understood as performing iterative inference (sequentially updating neural activity in order to improve the accuracy of beliefs). We propose a hybrid predictive coding network that combines both iterative and amortized inference in a principled manner by describing both in terms of a dual optimization of a single objective function. We show that the resulting scheme can be implemented in a biologically plausible neural architecture that approximates Bayesian inference utilising local Hebbian update rules. We demonstrate that our hybrid predictive coding model combines the benefits of both amortized and iterative inference—obtaining rapid and computationally cheap perceptual inference for familiar data while maintaining the context-sensitivity, precision, and sample efficiency of iterative inference schemes. Moreover, we show how our model is inherently sensitive to its uncertainty and adaptively balances iterative and amortized inference to obtain accurate beliefs using minimum computational expense. Hybrid predictive coding offers a new perspective on the functional relevance of the feedforward and recurrent activity observed during visual perception and offers novel insights into distinct aspects of visual phenomenology.Author summary: Predictive Coding Networks (PCNs) are a neurobiologically plausible model of cortical processing that can be applied to machine learning tasks. However, they require a computationally costly inference phase to generate predictions. We propose adding an amortized feedforward network to the model which learns to predict the outcome of iterative inference, and uses these predictions to initialize the predictive coding network—an approach we call hybrid predictive coding. This allows our hybrid model to perform simultaneous classification and generation and can be trained much faster and with less data than a standard PCN. Our model can also naturally and adaptively vary its computation time according to task demands, and may also help shed light on the neurocomputational basis of some otherwise difficult-to-understand aspects of visual phenomenology, thus suggesting that the brain may utilize a similar hybrid inference approach in visual processing.