Lossy Common Information in a Learnable Gray-Wyner Network
This work addresses inefficiencies in computer vision codecs by bridging classic information theory with modern machine learning, though it is incremental in applying Gray-Wyner theory to new contexts.
The paper tackled the problem of redundant representations in computer vision tasks by developing a learnable three-channel codec based on the Gray-Wyner network to disentangle shared and task-specific information, demonstrating substantial redundancy reduction and consistent outperformance over independent coding across six benchmarks.
Many computer vision tasks share substantial overlapping information, yet conventional codecs tend to ignore this, leading to redundant and inefficient representations. The Gray-Wyner network, a classical concept from information theory, offers a principled framework for separating common and task-specific information. Inspired by this idea, we develop a learnable three-channel codec that disentangles shared information from task-specific details across multiple vision tasks. We characterize the limits of this approach through the notion of lossy common information, and propose an optimization objective that balances inherent tradeoffs in learning such representations. Through comparisons of three codec architectures on two-task scenarios spanning six vision benchmarks, we demonstrate that our approach substantially reduces redundancy and consistently outperforms independent coding. These results highlight the practical value of revisiting Gray-Wyner theory in modern machine learning contexts, bridging classic information theory with task-driven representation learning.