Bayesian Statistics (with Particular Focus on the Motor System)
Probabilistic inference; Statistical inference; Bayesian inference; Antonym: Frequentist statistics
As our sensors are not perfect and do not provide information about all the properties of the world, we are faced with sensory uncertainty. Moreover, our muscles also produce noisy outputs and so we are also faced with motor uncertainty. Bayesian statistics is the systematic way of dealing with uncertainty by expressing its many forms in terms of probability.
Description of the Theory
Bayesian statistics can be used to infer the states of variables that are not directly measured, a process called Bayesian inference. “For example, when we see a tennis ball (observed variable) and we know from experience how they usually fly (prior knowledge) then Bayesian statistics allows us to calculate how likely the ball will land at any given position on the field (unobserved variable).” The way inference is done is the following:
(i) The prior knowledgeabout the system is...
- 2.Freedman DA (1995) Some issues in the foundation of statistics. Found Sci 1:19–83Google Scholar
- 7.Kalman RE (1960) A new approach to linear filtering and prediction problems. J Basic Eng (ASME) 82D:35–45Google Scholar
- 9.MacKay DJC (2003) Information Theory, Inference, and Learning Algorithms. Cambridge University press, Cambridge, UK [also available online at http://www.inference.phy.cam.ac.uk/mackay/itila/book.html]