Replicate

Costs a few pennies, API image generation

Lecture 1

AI vs ML vs DL

Deep Learning is much more recent, and less broad than AI or ML

AI > ML > DL

ML

ML: field of study that gives the computer the ability to learn without being explicitly programmed

• program with examples, not instructions

DL

ML + Powerful GPUs + Lots of Data

• uses Neural Networks for best accuracy

Supervised Learning: the images are labeled

Unsupervised: unlabeled data set, grouped into clustering (a group of dog images, group based on visual similarities)

Reinforcement Learning: learn by trial-error through reward or punishment

Think about the: Inputs > ML Algorithms > Output

Types of ML Outputs:

• Regression: predicts a continuous value (price of a stock)
• Classification: predict a class/label (cat/dog)
• Clustering: most similar other examples (recommendation)