CHOPT : Automated Hyperparameter Optimization Framework for Cloud-Based Machine Learning Platforms

Abstract

Many hyperparameter optimization (HyperOpt) methods assume restricted computing resources and mainly focus on enhancing performance. Here we propose a novel cloud-based HyperOpt (CHOPT) framework which can efficiently utilize shared computing resources while supporting various HyperOpt algorithms. We incorporate convenient web-based user interfaces, visualization, and analysis tools, enabling users to easily control optimization procedures and build up valuable insights with an iterative analysis procedure. Furthermore, our framework can be incorporated with any cloud platform, thus complementarily increasing the efficiency of conventional deep learning frameworks. We demonstrate applications of CHOPT with tasks such as image recognition and question-answering, showing that our framework can find hyperparameter configurations competitive with previous work. We also show CHOPT is capable of providing interesting observations through its analysing tools

arXiv comment10 pages, 9 figures. arXiv admin note: text overlap with arXiv:1807.01774 by other authors

References in corpus (7)

• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• An overview of gradient descent optimization algorithms
• WaveNet: A Generative Model for Raw Audio
• Wide Residual Networks
• Bidirectional Attention Flow for Machine Comprehension
• Tune: A Research Platform for Distributed Model Selection and Training
• NSML: A Machine Learning Platform That Enables You to Focus on Your Models

Cited by in corpus (5)

• HyperTendril: Visual Analytics for User-Driven Hyperparameter Optimization of Deep Neural Networks
• VisEvol: Visual Analytics to Support Hyperparameter Search through Evolutionary Optimization
• Stage-based Hyper-parameter Optimization for Deep Learning
• Hippo: Taming Hyper-parameter Optimization of Deep Learning with Stage Trees
• M2FN: Multi-step Modality Fusion for Advertisement Image Assessment