DEEP LEARNING
ARTIFICIAL INTELLIGENCE
https://rocka.co
xergioalex
BIASED ALGORITHMS
"AI is the new electricity"
Andrew Ng
Field of A.I focused on systems that can learn autonomously.
Learning in this context means finding complex patterns in millions of data
Â
Allows to make future predictions based on behaviors or characteristics that have been seen in the data already stored
1
CAT
0
NON-CAT
?
Use historical data that are not labeled The purpose is to explore them to find the structure or the way to organize them.
?
?
?
They are the basis of all AI. They work like the neurons of our brain at least at the conceptual level.
PREDICTIONS
(Markov chains)
Learning at multiple levels, each hidden layer is responsible for recognizing different characteristics, and deliver them as input to the next layer.
VECTORIZATION
The art of getting rid of the for loops in the code
SIMD
Single Instruction
Multiple Data
It is a technique used to achieve parallelism at the data level.
CPU VS GPU
Forward propagation
Backward propagation
HOW DEEP LEARNING WORKS
Z = Wx + b
Logistic Regresion
activation
function
(Z)
prediction
Loss Function
Cost Function
NON CAT (0)
HOW DEEP LEARNING WORKS
GRADIENT DESCENT
(Optimizer function)
GRADIENT DESCENT
(Optimizer function)
f'(x) = 0
GRADIENT DESCENT
SPECTROGRAM
SPECTROGRAM
SPECTROGRAM
CONVOLUTIONAL NEURONAL NETWORKS
CONVOLUTIONAL NEURONAL NETWORKS
CONVOLUTIONAL NEURONAL NETWORKS
https://www.deeplearning.ai/
CONVOLUTIONAL NEURONAL NETWORKS
CONVOLUTIONAL NEURONAL NETWORKS
CONVOLUTIONAL NEURONAL NETWORKS
Sara
COMPUTER VISION
COMPUTER VISION
https://rocka.co
xergioalex