Inferring Latent Functions with Gaussian Processes in Differential Equations

Posted in Science on August 23, 2008


Inferring Latent Functions with Gaussian Processes in Differential Equations

Lecture slides:

  • Inferring Latent Functions with Gaussian Processes in Differential Equations
  • Advert!
  • Gaussian Processes
  • Application Overview
  • Methodology
  • Linear Response Model
  • Linear Response Solution
  • Covariance Functions - Visualisation of RBF Covariance
  • Covariance Samples
  • Induced Covariance
  • Covariance Result
  • Cross Covariance
  • Posterior for f
  • Covariance for Transcription Model
  • Joint Sampling of x (t) and f (t)
  • Joint Sampling of x (t) and f (t)01
  • Joint Sampling of x (t) and f (t)02
  • Covariance for Transcription Model
  • Noise Corruption
  • Artificial Data Results
  • Results
  • Linear response analysis
  • Linear Response Results
  • Results Transcription Rates
  • Results Transcription Rates01
  • Results Transcription Rates02
  • Linear Response Discussion
  • Non-linear Response Model
  • Formalism
  • Example: linear response
  • Oscillatory Behaviour - Fix with MLP Covariance
  • Covariance Samples
  • Covariance Samples01
  • Response Results
  • Non-linear response analysis
  • exp (·) Response Results
  • log (1 + exp (f )) Response Results
  • 3/1+exp(-f ) Response Results
  • Discussion
  • Other Relevant Work
  • Open Question

Author: Neil Lawrence, University Of Manchester

Watch Video

Tags: Science, Lectures, Computer Science, Machine Learning, VideoLectures.Net, Gaussian Processes