1.1 - DL Overview

!wget -nc --no-cache -O init.py -q https://raw.githubusercontent.com/rramosp/2021.deeplearning/main/content/init.py
import init; init.init(force_download=False); 
from IPython.display import Image
Image(filename='local/imgs/ai_ml_dl.jpeg')
../_images/U1.01 - DL Overview_2_0.jpg
Image(filename='local/imgs/DL_timeline.png')
../_images/U1.01 - DL Overview_3_0.png

Some types of Neural Networks

http://www.asimovinstitute.org/neural-network-zoo/

Image(filename='local/imgs/nntypes.png')
../_images/U1.01 - DL Overview_5_0.png

Why DL now?

A “weird” introduction to Deep Learning

    As I said before, until the late 2000s, we were still missing a reliable way to train
    very deep neural networks. Nowadays, with the development of several simple but important
    theoretical and algorithmic improvements, the advances in hardware (mostly GPUs, now TPUs), 
    and the exponential generation and accumulation of data, DL came naturally to fit this 
    missing spot to transform the way we do machine learning.

Feature learning

Image(filename='local/imgs/feature_learning_ml_dl.png')
../_images/U1.01 - DL Overview_9_0.png

DL is suited for large datasets with highly dimensional input variables

Image(filename='local/imgs/cnn_feature_hierarchy.png')
../_images/U1.01 - DL Overview_11_0.png