BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Statistically robust\, scalable and distributed inference methods 
 for large scale data
DTSTART:20190502T133000
DTEND:20190502T143000
DTSTAMP:20260407T010611Z
UID:494fcb7e15af77e13969dfc91cea3578f1b556dcc0f8fa93c488df57
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Visa Koivunen\, Aalto University\, Finland\nIn this talk
  we address the problem of performing statistical inference for large scal
 e data sets. The volume and dimensionality of the data may be so high that
  it cannot be processed or stored in a single computing node. First\, we p
 resent a scalable\, statistically robust and computationally efficient bo
 otstrap method\, compatible with distributed processing and storage system
 s. Bootstrap resamples are constructed with smaller number of distinct dat
 a points on multiple disjoint subsets of data\, similarly to the bag of li
 ttle bootstrap method (BLB). A computationally efficient fixed-point est
 imation equation is analytically solved via a smart approximation stemming
  from the Fast and Robust Bootstrap method (FRB). Fixed point estimation e
 quations lend themselves to highly robust and low complexity statistical e
 stimators in finding point estimates\, confidence intervals and performing
  variable selection for large scale data sets. Sparse solutions can be pro
 moted\, too. We also propose a method for performing inference on fields o
 bserved by a massive number of spatially distributed sensors in IoT. The a
 pproach is nonparametric in a sense that it learns the probability models.
  The actual inference is based on fusing p-values and multiple hypothesis 
 testing controlling false discovery rates. A field may is clustered in hom
 ogeneous regions using an empirical Bayesian approach that takes the under
 lying spatial dependencies among the sensors into account. The clustering 
 finds applications in characterizing radio spectrum\, environmental monito
 ring\, cyber-physical systems and agriculture.
LOCATION:ELG 120 https://plan.epfl.ch/?room=ELG120
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR