adam-atan2-pytorch

Adam-atan2 - Pytorch

Implementation of the proposed Adam-atan2 optimizer in Pytorch

A multi-million dollar paper out of google deepmind proposes a small change to Adam update rule (using atan2) to remove the epsilon altogether for numerical stability and scale invariance

It also contains some features for improving plasticity (continual learning field)

Install

$ pip install adam-atan2-pytorch

Usage

import torch
from torch import nn

# toy model

model = nn.Linear(10, 1)

# import AdamAtan2 and instantiate with parameters

from adam_atan2_pytorch import AdamAtan2

opt = AdamAtan2(model.parameters(), lr = 1e-4)

# forward and backwards

for _ in range(100):
  loss = model(torch.randn(10))
  loss.backward()

  # optimizer step

  opt.step()
  opt.zero_grad()

Citations

@inproceedings{Everett2024ScalingEA,
    title   = {Scaling Exponents Across Parameterizations and Optimizers},
    author  = {Katie Everett and Lechao Xiao and Mitchell Wortsman and Alex Alemi and Roman Novak and Peter J. Liu and Izzeddin Gur and Jascha Narain Sohl-Dickstein and Leslie Pack Kaelbling and Jaehoon Lee and Jeffrey Pennington},
    year    = {2024},
    url     = {https://api.semanticscholar.org/CorpusID:271051056}
}

@inproceedings{Kumar2023MaintainingPI,
    title   = {Maintaining Plasticity in Continual Learning via Regenerative Regularization},
    author  = {Saurabh Kumar and Henrik Marklund and Benjamin Van Roy},
    year    = {2023},
    url     = {https://api.semanticscholar.org/CorpusID:261076021}
}

@article{Lewandowski2024LearningCB,
    title   = {Learning Continually by Spectral Regularization},
    author  = {Alex Lewandowski and Saurabh Kumar and Dale Schuurmans and Andr'as Gyorgy and Marlos C. Machado},
    journal = {ArXiv},
    year    = {2024},
    volume  = {abs/2406.06811},
    url     = {https://api.semanticscholar.org/CorpusID:270380086}
}