Michael Andersch

Artificial Intelligence · 2022

A researcher in the Science corpus: 1 paper; trusts 4 researchers and is trusted by 0 researchers here, as inferred from the citation graph.

Subfields

Whom Michael Andersch trusts overall

Their strongest outbound trust, strongest first, inferred from their citations.

  1. Samyam Rajbhandari

Papers from Michael Andersch's perspective

Ordered by how strongly the researchers Michael Andersch trusts endorse each paper. Same corpus, a different order for every researcher.

  1. Language Models are Few-Shot Learners (Neural Information Processing Systems)
  2. Training Deep Nets with Sublinear Memory Cost
  3. GPipe: Efficient Training of Giant Neural Networks using Pipeline Parallelism
  4. ZeRO-Offload: Democratizing Billion-Scale Model Training
  5. Mesh-TensorFlow: Deep Learning for Supercomputers
  6. Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
  7. PipeDream: generalized pipeline parallelism for DNN training
  8. ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep learning
  9. Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM
  10. DeepSpeed: System Optimizations Enable Training Deep Learning Models with Over 100 Billion Parameters

Their papers