Event Classification with Multi-step Machine Learning
This work addresses the challenge of efficiently designing modular ML systems for tasks like event classification, but it is incremental as it builds on existing NAS methods.
The paper tackles the problem of optimizing multi-step machine learning systems by using Neural Architecture Search (DARTS and SPOS-NAS) to select and connect pre-optimized models for sub-tasks, resulting in quick selection of high-performing model combinations that are consistent with baseline algorithms like grid search.
The usefulness and value of Multi-step Machine Learning (ML), where a task is organized into connected sub-tasks with known intermediate inference goals, as opposed to a single large model learned end-to-end without intermediate sub-tasks, is presented. Pre-optimized ML models are connected and better performance is obtained by re-optimizing the connected one. The selection of an ML model from several small ML model candidates for each sub-task has been performed by using the idea based on Neural Architecture Search (NAS). In this paper, Differentiable Architecture Search (DARTS) and Single Path One-Shot NAS (SPOS-NAS) are tested, where the construction of loss functions is improved to keep all ML models smoothly learning. Using DARTS and SPOS-NAS as an optimization and selection as well as the connections for multi-step machine learning systems, we find that (1) such a system can quickly and successfully select highly performant model combinations, and (2) the selected models are consistent with baseline algorithms, such as grid search, and their outputs are well controlled.