Main Menu (Mobile)- Block
- Overview
-
Support Teams
- Overview
- Anatomy and Histology
- Cryo-Electron Microscopy
- Electron Microscopy
- Flow Cytometry
- Gene Targeting and Transgenics
- Immortalized Cell Line Culture
- Integrative Imaging
- Invertebrate Shared Resource
- Janelia Experimental Technology
- Mass Spectrometry
- Media Prep
- Molecular Genomics
- Primary & iPS Cell Culture
- Project Pipeline Support
- Project Technical Resources
- Quantitative Genomics
- Scientific Computing Software
- Scientific Computing Systems
- Viral Tools
- Vivarium
- Open Science
- You + Janelia
- About Us
Main Menu - Block
- Overview
- Anatomy and Histology
- Cryo-Electron Microscopy
- Electron Microscopy
- Flow Cytometry
- Gene Targeting and Transgenics
- Immortalized Cell Line Culture
- Integrative Imaging
- Invertebrate Shared Resource
- Janelia Experimental Technology
- Mass Spectrometry
- Media Prep
- Molecular Genomics
- Primary & iPS Cell Culture
- Project Pipeline Support
- Project Technical Resources
- Quantitative Genomics
- Scientific Computing Software
- Scientific Computing Systems
- Viral Tools
- Vivarium

Abstract
An agglomerative clustering algorithm merges the most similar pair of clusters at every iteration. The function that evaluates similarity is traditionally handdesigned, but there has been recent interest in supervised or semisupervised settings in which ground-truth clustered data is available for training. Here we show how to train a similarity function by regarding it as the action-value function of a reinforcement learning problem. We apply this general method to segment images by clustering superpixels, an application that we call Learning to Agglomerate Superpixel Hierarchies (LASH). When applied to a challenging dataset of brain images from serial electron microscopy, LASH dramatically improved segmentation accuracy when clustering supervoxels generated by state of the boundary detection algorithms. The naive strategy of directly training only supervoxel similarities and applying single linkage clustering produced less improvement.