Discovering Multiple Algorithm Configurations

Many practitioners in robotics regularly depend on classic, hand-designed algorithms. Often the performance of these algorithms is tuned across a dataset of annotated examples which represent typical deployment conditions. Automatic tuning of these settings is traditionally known as algorithm configuration. In this work, we extend algorithm configuration to automatically discover multiple modes in the tuning dataset. Unlike prior work, these configuration modes represent multiple dataset instances and are detected automatically during the course of optimization. We propose three methods for mode discovery: a post hoc method, a multistage method, and an online algorithm using a multi-armed bandit. Our results characterize these methods on synthetic test functions and in multiple robotics application domains: stereoscopic depth estimation, differentiable rendering, motion planning, and visual odometry. We show the clear benefits of detecting multiple modes in algorithm configuration space.

Authors

• Leonid Keselman
• Martial Hebert

Keywords

• Automation
• Autonomous systems
• Stereo image processing
• Estimation
• Rendering (computer graphics)
• Planning
• Tuning
• Multiple Configurations
• Algorithm Configuration
• Path Planning
• Robotic Applications
• Online Algorithm
• Example Of Dataset
• Synthetic Function
• Multi-armed Bandit
• Visual Odometry
• Optimal Course
• Evaluation Of Function
• Feature Space
• Depth Camera
• Single Configuration
• Optimal Partition
• Machine Learning Community
• Domain-specific Features
• Black-box Optimization