Professional Documents
Culture Documents
Dlsportsterry150622 150624054011 Lva1 App6892 PDF
Dlsportsterry150622 150624054011 Lva1 App6892 PDF
University of Waterloo
Department of Electrical & Computer Engineering
Terry Taewoong Um
Weight
(kg)
[Output Y]
? Height(cm)
[Input X]
Weight Y = aX+b
(kg)
80
180 Height(cm)
Model : Y = aX+b Parameter : (a, b)
[Goal] Find (a,b) which best fits the given data
Terry Taewoong Um (terry.t.um@gmail.com)
11
A TO Y E X A M P L E
[Numerical Solution]
1. Set a cost function
Running
Record
(min)
140
32 Age(year)
“overfitting”
? X
• Another interpretation :
: Maximum a Posteriori (MAP)
http://goo.gl/6GE2ix
http://goo.gl/6GE2ix
• Another interpretation :
: Maximum a Posteriori (MAP)
http://goo.gl/6GE2ix
http://goo.gl/6GE2ix
error
test error
training validation test
set set set
for training for early for evaluation
training error (parameter stopping (measure the
optimization) (avoid performance)
we should overfitting)
training time
stop here
keep watching the validation error
kernel function
[Dual formulation] ( )
kernel function
likelihood
prior
posterior
https://youtu.be/kvPmArtVoFE
prediction conditioning the joint distribution of the observed & predicted values
https://youtu.be/YqhLnCm0KXY
http://goo.gl/XvOOmf
𝑋 → ∅(𝑋)
high dim. low dim.
Y = P X
http://jbhuang0604.blogspot.kr/2013/04/miss-korea-2013-contestants-face.html
PAST PRESENT
Knowledge Knowledge
PRESENT
kernel function
Representation
Knowledge
PRESENT
IMAGE
Knowledge
PRESENT
IMAGE
Representation
(Features)
Hand-Crafted Features
Can we learn a good representation
(feature) for the target task as well?
SPEECH
Knowledge
http://goo.gl/zNbBE2 http://goo.gl/Lk64Q4
- A single hidden layer NN w/ linear output can approximate any cont. func. arbitrarily well,
given enough hidden units
- This does not imply we have learning method to train them
• Activation functions
http://goo.gl/qMQk5H
: Backpropagation algorithm
http://goo.gl/qMQk5H
: Backpropagation algorithm
• Parameter gradients
known
http://goo.gl/qMQk5H
: Backpropagation algorithm
• Parameter gradients
known
http://goo.gl/qMQk5H
: Backpropagation algorithm
• Parameter gradients
known
http://goo.gl/qMQk5H
• NN is often likely to be
stuck in local minima pitfall
http://goo.gl/qMQk5H
Terry Taewoong Um (terry.t.um@gmail.com)
40
F R O M N N TO D E E P N N
• A long winter of NN
- NN requires expert’s skill to tune the hyperparameters
- It sometimes gives a good result, but sometimes gives a bad result.
The result is highly depend on the quality of initialization, regularization,
hyperparameters, data, etc.
http://t-robotics.blogspot.kr/2015/05/deep-learning.html
“Convolutional deep belief networks for scalable unsupervised learning of hierarchical representation”, Lee et al., 2012
https://goo.gl/Xswsbd
t-1 t t+1
http://www.dmi.usherb.ca/~larocheh/index_en.html
back propagation
through time
(BPTT)
back propagation
http://t-robotics.blogspot.kr/2015/05/deep-learning.html
3. Machine Learning in
Motion Analysis
TASKS
• Learning the kinematic/dynamic model
• Motion segmentation
• Motion generation / synthesis
• Motion imitation (Imitation learning)
• Activity / Gesture recognition
http://goo.gl/gFOVWL
Data Applications
• Motion capture data • Biomechanics
• Vision Data • Humanoid
• Dynamic-level data • Animation
- 3D convolution
“Hierarchical Recurrent Neural Network for Skeleton Based Action Recognition”, Y. Du et al., 2015
https://www.youtube.com/watch?v=oxA2O-tHftI