Classification and Reconstruction Processes in Deep Predictive Coding Networks: Antagonists or Allies?

TL;DR

This paper critically examines how classification and reconstruction processes interact in deep predictive coding networks, revealing a trade-off where each process diminishes the other's information in shared representations, challenging common assumptions.

Contribution

It provides a systematic analysis of the interaction between classification and reconstruction in predictive coding-inspired networks, highlighting the inherent trade-offs and potential ways to mitigate them.

Findings

Classification reduces reconstruction information in shared layers.

Increasing network complexity can alleviate the trade-off.

Shared representation dimensions impact the coexistence of processes.

Abstract

Predictive coding-inspired deep networks for visual computing integrate classification and reconstruction processes in shared intermediate layers. Although synergy between these processes is commonly assumed, it has yet to be convincingly demonstrated. In this study, we take a critical look at how classifying and reconstructing interact in deep learning architectures. Our approach utilizes a purposefully designed family of model architectures reminiscent of autoencoders, each equipped with an encoder, a decoder, and a classification head featuring varying modules and complexities. We meticulously analyze the extent to which classification- and reconstruction-driven information can seamlessly coexist within the shared latent layer of the model architectures. Our findings underscore a significant challenge: Classification-driven information diminishes reconstruction-driven information in intermediate layers' shared representations and vice versa. While expanding the shared representation's dimensions or increasing the network's complexity can alleviate this trade-off effect, our results challenge prevailing assumptions in predictive coding and offer guidance for future iterations of predictive coding concepts in deep networks.