shreyansh26 / ml-optimizers-jax

Toy implementations of some popular ML optimizers using Python/JAX

ML Optimizers from scratch using JAX

Implementations of some popular optimizers from scratch for a simple model i.e., Linear Regression on a dataset of 5 features. The goal of this project was to understand how these optimizers work under the hood and try to do a toy implementation myself. I also use a bit of JAX magic to perform the differentiation of the loss function w.r.t to the weights and the bias without explicitly writing their derivatives as a separate function. This can help to generalize this notebook for other types of loss functions as well.

The optimizers I have implemented are -

• Batch Gradient Descent
• Batch Gradient Descent + Momentum
• Nesterov Accelerated Momentum
• Adagrad
• RMSprop
• Adam
• Adamax
• Nadam
• Adabelief

References -

• https://ruder.io/optimizing-gradient-descent/
• https://theaisummer.com/optimization/