stream flow rate prediction

4
Stream flow rate prediction x = weather data f(x) = flow rate 11 12 13 14 15 16 0 1 2 3 4 5 6 Years ofD ata Transfered M A E (m ^3/s) N o Transfer W ilmot S harpe S harpe & W ilmot S hubenacadie

Upload: caryn-weber

Post on 31-Dec-2015

74 views

Category:

Documents


1 download

DESCRIPTION

x = weather data f(x ) = flow rate. Stream flow rate prediction. Domain Knowledge long-term memory. Retention & Consolidation. Knowledge Transfer. Inductive Bias. Selection. Transfer Learning and MTL. Testing Examples. Instance Space X. ( x, f ( x )). Model of Classifier h. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Stream flow rate  prediction

Stream flow rate prediction

x = weather data

f(x) = flow rate

11

12

13

14

15

16

0 1 2 3 4 5 6Years of Data Transfered

MA

E (

m^

3/s)

No Transfer Wilmot Sharpe Sharpe & Wilmot Shubenacadie

Page 2: Stream flow rate  prediction

2

Transfer Learning and MTL

Instance Space X

TrainingExamples

TestingExamples

(x, f(x))

Model ofClassifier

h

Inductive Learning Systemshort-term memory

h(x) ~ f(x)

DomainKnowledge

long-term memoryRetention &ConsolidationInductive

Bias SelectionKnowledgeTransfer

Page 3: Stream flow rate  prediction

3

Transfer Learning and MTL

Instance Space X

TrainingExamples

TestingExamples

(x, f(x))

Model ofClassifier

h

h(x) ~ f(x)

DomainKnowledge

long-term memoryRetention &ConsolidationKnowledge

Transfer

f2(x)

x1 xn

f1(x) fk(x)

Multiple Task Learning (MTL)

InductiveBias Selection

Page 4: Stream flow rate  prediction

4

Multiple Task Learning (MTL)

f2(x)

x1 xn

f1(x)

Task specificrepresentation

Common internalRepresentation

Common feature layer

fk(x)

Multiple hypotheses develop in parallel within one back-propagation network [Caruana, Baxter, Silver 93-95]

An inductive bias occurs through shared use of common internal representation

Knowledge or Inductive transfer to primary task f1 (x) depends on choice of secondary tasks