More Than A Shortcut: A Hyperbolic Approach To Early-Exit Networks
By: Swapnil Bhosale , Cosmin Frateanu , Camilla Clark and more
Potential Business Impact:
Makes smart devices detect sounds more accurately.
Deploying accurate event detection on resource-constrained devices is challenged by the trade-off between performance and computational cost. While Early-Exit (EE) networks offer a solution through adaptive computation, they often fail to enforce a coherent hierarchical structure, limiting the reliability of their early predictions. To address this, we propose Hyperbolic Early-Exit networks (HypEE), a novel framework that learns EE representations in the hyperbolic space. Our core contribution is a hierarchical training objective with a novel entailment loss, which enforces a partial-ordering constraint to ensure that deeper network layers geometrically refine the representations of shallower ones. Experiments on multiple audio event detection tasks and backbone architectures show that HypEE significantly outperforms standard Euclidean EE baselines, especially at the earliest, most computationally-critical exits. The learned geometry also provides a principled measure of uncertainty, enabling a novel triggering mechanism that makes the overall system both more efficient and more accurate than a conventional EE and standard backbone models without early-exits.
Similar Papers
AEBNAS: Strengthening Exit Branches in Early-Exit Networks through Hardware-Aware Neural Architecture Search
Artificial Intelligence
Makes smart computer programs use less power.
Hetero-SplitEE: Split Learning of Neural Networks with Early Exits for Heterogeneous IoT Devices
Machine Learning (CS)
Lets different smart devices train AI together.
Boosted Training of Lightweight Early Exits for Optimizing CNN Image Classification Inference
CV and Pattern Recognition
Makes AI recognize pictures faster and with less power.