Header logo is ps

Active Learning for Abstract Models of Collectives


Conference Paper


Organizational structures such as hierarchies provide an effective means to deal with the increasing complexity found in large-scale energy systems. In hierarchical systems, the concrete functions describing the subsystems can be replaced by abstract piecewise linear functions to speed up the optimization process. However, if the data points are weakly informative the resulting abstracted optimization problem introduces severe errors and exhibits bad runtime performance. Furthermore, obtaining additional point labels amounts to solving computationally hard optimization problems. Therefore, we propose to apply methods from active learning to search for informative inputs. We present first results experimenting with Decision Forests and Gaussian Processes that motivate further research. Using points selected by Decision Forests, we could reduce the average mean-squared error of the abstract piecewise linear function by one third.

Author(s): Alexander Schiendorfer and Christoph Lassner and Gerrit Anders and Wolfgang Reif and Rainer Lienhart
Book Title: 3rd Workshop on Self-optimisation in Organic and Autonomic Computing Systems (SAOS)
Year: 2015
Month: March

Department(s): Perceiving Systems
Bibtex Type: Conference Paper (inproceedings)
Paper Type: Conference

Event Place: Porto

Links: code (hosted on github)
Attachments: pdf


  title = {Active Learning for Abstract Models of Collectives},
  author = {Schiendorfer, Alexander and Lassner, Christoph and Anders, Gerrit and Reif, Wolfgang and Lienhart, Rainer},
  booktitle = {3rd Workshop  on Self-optimisation in Organic and Autonomic Computing Systems (SAOS)},
  month = mar,
  year = {2015},
  month_numeric = {3}