11 papers
Phases of Muon: When Muon Eclipses SignSGD
Elliot Paquette, Noah Marshall, Lucas Benigni +3
Recently, Muon and related spectral optimizers have demonstrated strong empirical performance as scalable stochastic methods, often outperforming Adam. Yet their behaviour remains…
High dimensional theory of two-phase optimizers
Atish Agarwala
The trend towards larger training setups has brought a renewed interest in partially asynchronous two-phase optimizers which optimize locally and then synchronize across workers. A…
Exact Risk Curves of signSGD in High-Dimensions: Quantifying Preconditioning and Noise-Compression Effects
Ke Liang Xiao, Noah Marshall, Atish Agarwala +1
In recent years, signSGD has garnered interest as both a practical optimizer as well as a simple model to understand adaptive optimizers like Adam. Though there is a general consen…
What do near-optimal learning rate schedules look like?
Hiroki Naganuma, Atish Agarwala, Priya Kasimbeg +1
A basic unanswered question in neural network training is: what is the best learning rate schedule shape for a given workload? The choice of learning rate schedule is a key factor…
Per-example gradients: a new frontier for understanding and improving optimizers
Vincent Roulet, Atish Agarwala
Training algorithms in deep learning usually treat a mini-batch of samples as a single object; they average gradients over the mini-batch, and then process the average in various w…
Scaling Collapse Reveals Universal Dynamics in Compute-Optimally Trained Neural Networks
Shikai Qiu, Lechao Xiao, Andrew Gordon Wilson +2
What scaling limits govern neural network training dynamics when model size and training time grow in tandem? We show that despite the complex interactions between architecture, tr…