34 citations · 49 across the 12 of their papers we have counts for
12 papers · 1 filter
Improving Neural Network Training by Decoupling the Magnitude and Direction of Weight Vectors
Alexander Hägele, Alejandro Hernández-Cano, Atli Kosson +1
Modern neural network training relies on optimizers such as Adam and Muon which act on each weight matrix as a single object. Yet every weight matrix carries two distinct quantitie…
Weight Decay may matter more than muP for Learning Rate Transfer in Practice
Atli Kosson, Jeremy Welborn, Yang Liu +2
Transferring the optimal learning rate from small to large neural networks can enable efficient training at scales where hyperparameter tuning is otherwise prohibitively expensive.…
Training Dynamics of the Cooldown Stage in Warmup-Stable-Decay Learning Rate Scheduler
Aleksandr Dremov, Alexander Hägele, Atli Kosson +1
Learning rate scheduling is essential in transformer training, where the final annealing plays a crucial role in getting the best performance. However, the mechanisms behind this c…
Analyzing & Reducing the Need for Learning Rate Warmup in GPT Training
Atli Kosson, Bettina Messmer, Martin Jaggi
Learning Rate Warmup is a popular heuristic for training neural networks, especially at larger batch sizes, despite limited understanding of its benefits. Warmup decreases the upda…
Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations
Alexander Hägele, Elie Bakouch, Atli Kosson +3
Scale has become a main ingredient in obtaining strong machine learning models. As a result, understanding a model's scaling properties is key to effectively designing both the rig…
Memory Efficient Mixed-Precision Optimizers
Basile Lewandowski, Atli Kosson
Traditional optimization methods rely on the use of single-precision floating point arithmetic, which can be costly in terms of memory size and computing power. However, mixed prec…
Rotational Equilibrium: How Weight Decay Balances Learning Across Neural Networks
Atli Kosson, Bettina Messmer, Martin Jaggi
This study investigates how weight decay affects the update behavior of individual neurons in deep neural networks through a combination of applied analysis and experimentation. We…
Ghost Noise for Regularizing Deep Neural Networks
Atli Kosson, Dongyang Fan, Martin Jaggi
Batch Normalization (BN) is widely used to stabilize the optimization process and improve the test performance of deep neural networks. The regularization effect of BN depends on t…
Multiplication-Free Transformer Training via Piecewise Affine Operations
Atli Kosson, Martin Jaggi
Multiplications are responsible for most of the computational cost involved in neural network training and inference. Recent research has thus looked for ways to reduce the cost as…
Adaptive Braking for Mitigating Gradient Delay
Abhinav Venigalla, Atli Kosson, Vitaliy Chiley +1
Neural network training is commonly accelerated by using multiple synchronized workers to compute gradient updates in parallel. Asynchronous methods remove synchronization overhead…
Pipelined Backpropagation at Scale: Training Large Models without Batches
Atli Kosson, Vitaliy Chiley, Abhinav Venigalla +2
New hardware can substantially increase the speed and efficiency of deep neural network training. To guide the development of future hardware architectures, it is pertinent to expl…
Online Normalization for Training Neural Networks
Vitaliy Chiley, Ilya Sharapov, Atli Kosson +5
Online Normalization is a new technique for normalizing the hidden activations of a neural network. Like Batch Normalization, it normalizes the sample dimension. While Online Norma…