BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Accelerated Free-Form Model Discovery of Interpretable Models usin
 g Small Data - Lior Horesh (IBM Research)
DTSTART:20171031T111000Z
DTEND:20171031T120000Z
UID:TALK94111@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:The ability to abstract the behavior of a system or a phenomen
 on and distill it into a consistent mathematical model is instrumental for
  a broad range of applications. Historically\, models were manually derive
 d in a first principles fashion. The first principles approach often offer
 s the derivation of interpretable models of remarkable levels of universal
 ity using little data. Nevertheless\, their derivation is time consuming a
 nd relies heavily upon domain expertise. Conversely\, with the rising perv
 asiveness of data-driven approaches\, the rapid derivation and deployment 
 of models has become a reality. Scalability is gained through dependence u
 pon exploitable structure (functional form). Such structures\, in turn\, y
 ield non-interpretable models\, require Big Data for training\, and provid
 e limited predictive power outside the training set span. In this talk\, w
 e will introduce an accelerated model discovery approach that attempts to 
 bridge between the two conducts\, to enable the discovery of universal\, i
 nterpretable models\, using Small Data. To accomplish that\, the proposed 
 algorithm searches for free-form symbolic models\, where neither the struc
 ture nor the set of operator primitives are predetermined. The discovered 
 models are provably globally optimal\, promoting superior predictive power
  for unseen input. Demonstration of the algorithm in re-discovery of some 
 fundamental laws of science will be provided\, and references to on-going 
 work in the discovery of new models for\, hitherto\, unexplainable phenome
 na.  Globally optimal symbolic regression\, NIPS Interpretable ML Workshop
 \, 2017\, https://arxiv.org/abs/1710.10720 Globally optimal Mixed Integer 
 Non-Linear Programming (MINLP) formulation for symbolic regression\, IBM T
 echnical Report ID 219095\, 2016
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR