Multi-Exit DNN Inference Acceleration Based on Multi-Dimensional Optimization for Edge Intelligence

Edge intelligence, as a prospective paradigm for accelerating DNN inference, is mostly implemented by model partitioning which inevitably incurs the large transmission overhead of DNN’s intermediate data. A popular solution introduces multi-exit DNNs to reduce latency by enabling early exits. However, existing work ignores the correlation between exit settings and synergistic inference, causing incoordination of device-to-edge. To address this issue, this paper first investigates the bottlenecks of executing multi-exit DNNs in edge computing and builds a novel model for inference acceleration with exit selection, model partition, and resource allocation. To tackle the intractable coupling subproblems, we propose a
M
ulti-exit DNN inference
A
cceleration framework based on
M
ulti-dimensional
O
ptimization (MAMO). In MAMO, the exit selection subproblem is first extracted from the original problem. Then, bidirectional dynamic programming is employed to determine the optimal exit setting for an arbitrary multi-exit DNN. Finally, based on the optimal exit setting, a DRL-based policy is developed to learn joint decisions of model partition and resource allocation. We deploy MAMO on a real-world testbed and evaluate its performance in various scenarios. Extensive experiments show that it can adapt to heterogeneous tasks and dynamic networks, and accelerate DNN inference by up to
$13.7\times$
compared with the state-of-the-art.