Shane Bergsma

1. Don't be lazy: CompleteP enables compute-efficient deep transformers
   2025/05/02 by Dey, Nolan, Bin Zhang, Lorenzo Noci +14 · 1 voice · 21 citations
   Computer Science · #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Neural Networks and Applications #Parallel Computing and Optimization Techniques
2. Sparse maximal update parameterization: A holistic approach to sparse training dynamics
   2024/05/24 by Nolan Dey, Shane Bergsma, Dey, Nolan +3 · 1 voice · 2 citations
   Computer Science · Physics and Astronomy · #Advanced Vision and Imaging #FOS: Computer and information sciences #Machine Learning (cs.LG) #Model Reduction and Neural Networks #Speech and Audio Processing #cs.LG
3. Normalization Layer Per-Example Gradients are Sufficient to Predict Gradient Noise Scale in Transformers
   2024/11/01 by G. T. Gray, Gray, Gavia, Tiwari, Aman +4 · 2 citations
   Engineering · Materials Science · #Advancements in Semiconductor Devices and Circuit Design #FOS: Computer and information sciences #I.2.6 #Integrated Circuits and Semiconductor Failure Analysis #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Magnetic Properties and Applications