Deep Learning Machine Learning in der Medizin Asan Agibetov, PhD [email protected]Medical University of Vienna Center for Medical Statistics, Informatics and Intelligent Systems Institute for Artificial Intelligence and Decision Support Spitalgasse 23, 1090 Vienna, BT88.04.808 December 13, 2018
29
Embed
Deep Learning - Machine Learning in der Medizin · 2018-12-17 · Why Deep Learning Peter Norvig’s 1 recollection on Geoff Hinton’s2 talk on Boltzmann Machine 3 work (back in
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Medical University of ViennaCenter for Medical Statistics, Informatics and Intelligent Systems
Institute for Artificial Intelligence and Decision SupportSpitalgasse 23, 1090 Vienna, BT88.04.808
December 13, 2018
Introduction
▶ References (available online for free):▶ ”Neural Networks and Deep Learning”. Michael A. Nielsen,
Determination Press, 2015▶ intuition first, math after
▶ ”Deep Learning”. Ian Goodfellow, Youshua Bengio and AaronCourville, MIT Press, 2016
▶ formal with fair amount of intuition, more general than Nielsen
▶ These slides are based on DL courses:▶ Course notes ”CNN for Visual Recognition” (Stanford, Spring
2017)▶ Course notes ”An introduction to Deep Learning”,
Marc’Aurelio Ranzato (Facebook AI Research), DeepLearnSummer School - Bilbao, 17-21 July 2017
Why Deep Learning
Peter Norvig’s 1 recollection on Geoff Hinton’s 2 talk onBoltzmann Machine 3 work (back in 1980)
1. Cognitive plausibility in terms of a model of the brain2. Model that learns from experiences rather that programmed by
hand3. Continuous representations rather than Boolean, as in
traditional symbolic expert systems
1Research Director at Google, co-author of classical texts on AI2Professor at University of Toronto, one of the pioneers of Deep Learning3Boltzmann Machine (and Probabilistic Graphical Models) one of the
theoretical foundations for generative DL models
Neural networks and Deep Learning
▶ Neural networks - biologically-inspired programming paradigm▶ enables computer to learn from observational data▶ universal function approximation machine 4
▶ Deep learning - powerful set of techniques for learning inneural networks
▶ harness GPU resources to parallelize and speed upmatrix-vector computations
▶ give rise to modularized approach to learning
4Hornik, ”Approximation capabilities of Multilayer Feedforward Networks”,Neural Networks, 1991
Deep Learning - what’s in the name?▶ DL, roughly speaking, is NN with many layers and many
neurons in each layer▶ not true in all cases though (e.g., embeddings are often
Why non linear layers▶ ReLU layers provide piece-wise linear tiling▶ # planes grows exponentially w. # hidden units▶ Multiple layers yield exponential savings in # parameters
(parameter sharing)
Figure 6: with ReLU mapping is locally linear 10
10Montufar et al. ”On the number of linear regions of DNNs”, arXiv, 2014
How good is the network: task-dependant loss function Vi