A Structure-Aware Framework for Learning Device Placements on Computation Graphs

TL;DR

This paper introduces a novel structure-aware framework for device placement on computation graphs, improving neural network inference speed by up to 60%, by combining graph coarsening, representation learning, and reinforcement learning.

Contribution

It bridges encoder-placer and grouper-placer methods, proposing a flexible, end-to-end trainable framework that considers the DAG structure for optimized device placement.

Findings

Achieves up to 58.2% speedup on CPU inference.

Demonstrates effectiveness on models like Inception-V3, ResNet, and BERT.

Shows robustness through ablation studies.

Abstract

Computation graphs are Directed Acyclic Graphs (DAGs) where the nodes correspond to mathematical operations and are used widely as abstractions in optimizations of neural networks. The device placement problem aims to identify optimal allocations of those nodes to a set of (potentially heterogeneous) devices. Existing approaches rely on two types of architectures known as grouper-placer and encoder-placer, respectively. In this work, we bridge the gap between encoder-placer and grouper-placer techniques and propose a novel framework for the task of device placement, relying on smaller computation graphs extracted from the OpenVINO toolkit. The framework consists of five steps, including graph coarsening, node representation learning and policy optimization. It facilitates end-to-end training and takes into account the DAG nature of the computation graphs. We also propose a model variant, inspired by graph parsing networks and complex network analysis, enabling graph representation learning and jointed, personalized graph partitioning, using an unspecified number of groups. To train the entire framework, we use reinforcement learning using the execution time of the placement as a reward. We demonstrate the flexibility and effectiveness of our approach through multiple experiments with three benchmark models, namely Inception-V3, ResNet, and BERT. The robustness of the proposed framework is also highlighted through an ablation study. The suggested placements improve the inference speed for the benchmark models by up to 58.2% over CPU execution and by up to 60.24% compared to other commonly used baselines.