BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//jEvents 2.0 for Joomla//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:eu7itt9mmoq44eblurhv2a0g8s@google.com
CATEGORIES:RTportal.ru
CREATED:20170606T092329
SUMMARY:Коллоквиум ФКН: Causal inference and Kolmogorov complexity
LOCATION:Кочновский проезд, д. 3, лекционный зал Декарт, 3 этаж.
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:t is often stated that “Correlation does not imply causation”: a dependency
  between observed values of two random variables might not imply that there
  exists a causal connection between the corresponding processes (and assumi
 ng there exists one, one might not know the direction). In Shannon informat
 ion theory, this is reflected by the law of “symmetry of information”: I(X;
 Y) = I(Y;X). The information that a random variable X has about Y equals th
 e information that Y has about X.<br/><br/>This law remains valid if Shanno
 n entropy is replaced by Kolmogorov complexity. However, there exists a sub
 tler setting where this law is violated and one might speculate that this a
 symmetry can be used to reconstruct causality.<br/><br/>In the second part 
 of the talk, we discuss the postulate of independence of conditionals for t
 he inference of causal relations in observed data. This postulate was intro
 duced by D. Janzing and B. Schölkopf in 2010, and the two-variable case sta
 tes that, if X causes Y, then the marginal distribution P X  has low inform
 ation about the conditional distribution P {Y|X}  (as defined with Kolmogor
 ov complexity). We explain how the most popular methods can be explained by
  this postulate and overview new methods that were inspired by it. This eve
 nt was imported from: https://rtportal.ru/index.php/kalendar-sobytij/eventd
 etail/166/-/kollokvium-fkn-causal-inference-and-kolmogorov-complexity?tmpl=
 component
DTSTAMP:20260812T210635Z
DTSTART;TZID=Europe/Moscow:20160510T181000
DTEND;TZID=Europe/Moscow:20160510T194000
SEQUENCE:0
TRANSP:OPAQUE
END:VEVENT
END:VCALENDAR