@InProceedings{Pecka-MESAS2014,
  IS = { zkontrolovano 03 Jan 2015 },
  UPDATE  = { 2014-12-19 },
  author =      { Pecka, Martin and Svoboda, Tom{\'a}{\v s} },
  affiliation = {13133-13133},
  title =       {Safe Exploration Techniques for Reinforcement Learning -- An Overview},
  year =        {2014},
  pages =       {357-375},
  booktitle =   {Modelling and Simulation for Autonomous Systems},
  editor =      {Jan Hodicky},
  publisher =   {Springer},
  address =     {Cham, Switzerland },
  isbn =        {978-3-319-13822-0},
  volume =      {1},
  series =      {Lecture Notes in Computer Science},
  number =      {8906},
  book_pages =  {388},
  month =       {May},
  day =         {5-6},
  venue =       {Rome, Italy},
  organization ={NATO Modelling and Simulation Centre of Excellence},
  annote =      {We overview different approaches to safety in
                  (semi)autonomous robotics. Part icularly, we focus
                  on how to achieve safe behavior of a robot if it is
                  requested to perform ex ploration of unknown
                  states. Presented methods are studied from the
                  viewpoint of reinforcement learning, a
                  partially-supervised machine learning method. To
                  collect training data for this a lgorithm, the robot
                  is required to freely explore the state space -
                  which can lead to possibly dangerous situations. The
                  role of safe exploration is to provide a framework
                  allowing explora tion while preserving safety. The
                  examined methods range from simple algorithms to
                  sophisticat ed methods based on previous experience
                  or state prediction. Our overview also addresses the
                  i ssues of how to define safety in the real-world
                  applications (apparently absolute safety is un
                  achievable in the continuous and random real
                  world). In the conclusion we also suggest several
                  ways that are worth researching more thoroughly.},
  keywords =    {Safe exploration, policy search, reinforcement learning},
  project =     {SGS13/142/OHK3/2T/13, FP7-ICT-609763 TRADR},
  doi =         {10.1007/978-3-319-13823-7},
  psurl =       {[PDF, 2 000 kB] },
}