ESP: Rapid Prototyping of Machine Learning for Real-Time Swarm Applications
David Mellis, Ben Zhang, Audrey Leung, Bjoern Hartmann

Citation
David Mellis, Ben Zhang, Audrey Leung, Bjoern Hartmann. "ESP: Rapid Prototyping of Machine Learning for Real-Time Swarm Applications". Talk or presentation, 26, October, 2016; Poster presented at the 2016 TerraSwarm Annual Meeting.

Abstract
Our Example-based Sensor Predictions system (ESP) supports rapid prototyping of machine learning for real-time IoT sensor data. ESP users author domain-specific machine learning pipelines in C++ using a general-purpose machine learning library and provide domain-specific heuristics. The ESP interface then allows for rapid experimentation with these pipelines in the context of a larger swarm system. This includes collection of training data, tuning of pipeline parameters, and visualization of data and predictions. This supports the rapid prototyping of new interactive swam applications that make use of machine learning.

Electronic downloads

Internal. This publication has been marked by the author for TerraSwarm-only distribution, so electronic downloads are not available without logging in.

Citation formats

• HTML

  David Mellis, Ben Zhang, Audrey Leung, Bjoern Hartmann.
  <a
  href="http://www.terraswarm.org/pubs/869.html"><i>ESP:
  Rapid Prototyping of Machine Learning for Real-Time Swarm
  Applications</i></a>, Talk or presentation,  26,
  October, 2016; Poster presented at the <a
  href="http://terraswarm.org/conferences/16/annual"
  >2016 TerraSwarm Annual Meeting</a>.

• Plain text

  David Mellis, Ben Zhang, Audrey Leung, Bjoern Hartmann.
  "ESP: Rapid Prototyping of Machine Learning for
  Real-Time Swarm Applications". Talk or presentation, 
  26, October, 2016; Poster presented at the <a
  href="http://terraswarm.org/conferences/16/annual"
  >2016 TerraSwarm Annual Meeting</a>.

• BibTeX

  @presentation{MellisZhangLeungHartmann16_ESPRapidPrototypingOfMachineLearningForRealTimeSwarm,
      author = {David Mellis and Ben Zhang and Audrey Leung and
                Bjoern Hartmann},
      title = {ESP: Rapid Prototyping of Machine Learning for
                Real-Time Swarm Applications},
      day = {26},
      month = {October},
      year = {2016},
      note = {Poster presented at the <a
                href="http://terraswarm.org/conferences/16/annual"
                >2016 TerraSwarm Annual Meeting</a>.},
      abstract = {Our Example-based Sensor Predictions system (ESP)
                supports rapid prototyping of machine learning for
                real-time IoT sensor data. ESP users author
                domain-specific machine learning pipelines in C++
                using a general-purpose machine learning library
                and provide domain-specific heuristics. The ESP
                interface then allows for rapid experimentation
                with these pipelines in the context of a larger
                swarm system. This includes collection of training
                data, tuning of pipeline parameters, and
                visualization of data and predictions. This
                supports the rapid prototyping of new interactive
                swam applications that make use of machine
                learning.},
      URL = {http://terraswarm.org/pubs/869.html}
  }
  

Posted by David Mellis on 28 Oct 2016.
Groups: pw