m8ta
You are not authenticated, login.
text: sort by
tags: modified
type: chronology
[0] Fetz EE, Perlmutter SI, Prut Y, Functions of mammalian spinal interneurons during movement.Curr Opin Neurobiol 10:6, 699-707 (2000 Dec)

[0] Peters J, Schaal S, Reinforcement learning of motor skills with policy gradients.Neural Netw 21:4, 682-97 (2008 May)

{288}
hide / / print
ref: Fetz-2000.12 tags: motor control spinal neurons interneurons movement primitives Fetz review tuning date: 01-03-2012 23:08 gmt revision:4 [3] [2] [1] [0] [head]

PMID-11240278[0] Functions of mammalian spinal interneurons during movement

  • this issue of current opinion in neuro has many reviews of motor control
  • points out that the Bizzi results (they microstimulated & observed a force-field-primitive type organization)
    • others have found that this may be a consequence of decerebration + the structure of the biomechanical groupings of muscles. (see 'update').
  • intraspinal electrodes in the cat provide a secure and reliable method of eliciting forces and movements.
  • CM (corticomotor) cells more often represent synergistic groups of muscles, whereas premotor spinal interneurons are organized to target specific muscles.
    • CMs are therefore more strictly recruited for particular movements.
  • interneurons (IN) are, of course, arrayed in such a way so that antagonist and agonist muscles cross-inhibit eachother (for efficiency)
    • however, we are still able to control the endpoint impedance of the arm - how?
  • spinal interneurons modulate activity during wait period prior to movement!
    • there might be substantial interaction between the cortex and spinal cord.. subjects asked to imagine pressing a foot pedal showed enhanced reflexes in the involved soleus muscle.
      • cognitive priming?
  • spinal reflexes are strongly modulated in movement.

____References____

{651}
hide / / print
ref: Peters-2008.05 tags: Schaal reinforcement learning policy gradient motor primitives date: 02-17-2009 18:49 gmt revision:4 [3] [2] [1] [0] [head]

PMID-18482830[0] Reinforcement learning of motor skills with policy gradients

  • they say that the only way to deal with reinforcement or general-type learning in a high-dimensional policy space defined by parameterized motor primitives are policy gradient methods.
  • article is rather difficult to follow; they do not always provide enough details (for me) to understand exactly what their equations mean. Perhaps this is related to their criticism that others's papers are 'ad-hoc' and not 'statistically motivated'
  • none the less, it seems interesting..
  • their previous paper - Reinforcement learning for Humanoid robotics - maybe slightly easier to understand.

____References____

{141}
hide / / print
ref: learning-0 tags: motor control primitives nonlinear feedback systems optimization date: 0-0-2007 0:0 revision:0 [head]

http://hardm.ath.cx:88/pdf/Schaal2003_LearningMotor.pdf not in pubmed.