Deeplearning,transfer learning,broader view Flashcards
Kernel
Kernel function
Non-linear phenomena can be linearized using
a kernel function
* Non-linear mapping + linear model
* Easy to implement / relies on simple models
* Kernel is often application-dependent, hard to
generalize
DL – advantages
Handcrafted features might be incomplete
and require a lot of human work for designing
and validation
* Learned features are adapted to the input
data
* Learned representations work for both
supervised and non-supervised applications
* Data representation and classification into one
single network – end-to-end learning.
DL Actiavtion functions:
Sigmoid neurons saturate and
kill gradients
Tanh function:
* Saturation: similar to
sigmoid
* Output is zero-centered
* Tanh is a scaled sigmoid:
RELU
Faster training thanks to its
non-saturating nature
– Prevents the gradient
vanishing problem
* Lighter computation
Fully Connected feedforwardNetwork
Several layers
No loops
High nr of connections
Usage:Classification
FROM NN TO CNN
1-Local connectivity:
only neighbors are connected. lower number of weigts to train
2-Shared weights.
many units share parameters but on diferent input windows.
3-Multi feature maps
Compute different function on same input.
4-Subsampling(pooling)
Reduces resolution
reduce computional complexity
adds some deformation to invariance
max pooling fast effixient and strong.
CNN regularization
1-Dropout
Randomply drop units with fixed proability
2-Early Stopping.
Stop when validation error is not improved
3.Preprocessing
add noise rotation flip etx data augmentation.
4.Batch processing because gradient is slow.
Video Sequences
-Optical Flow
2D motion of image between consecutive images.
3D motion to 2 D.
-Motion flow: actual 2d motion of points.
Beyond Natural Lighting
Lambertian material: Perfect difuse
EM spectrum
much wider than visible specturm
Multispectral imaging
-wider portion of em specturm
-multiple bands
Hyperspectral
-wider portion of em spectrum
-fine grained sample
-Composed of a huge number of channels
-Measure “chemical”-related properties of
materials
-Enable multiple analysis to be performed with one
single image