Ethan Dyer
9 papers · 2020–2022 · 3 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+3 more ↓ Show less ↑
π£ Hot Topic Early Bird π§ Keyword Pioneer π Conference Polyglot (3) π Cross-Pollinator (6) π Renaissance Researcher (5)
π
Interdisciplinary Bridge
πΊοΈ
Taxonomy Completionist
(17)
β
The Questioner
Conferences
ICLR (5)
NIPS (3)
ICML (1)
Top co-authors
Keywords
information theory
(1)
sequence modeling
(1)
neural network training
(1)
mathematical reasoning
(1)
in-context learning
(1)
language modeling
(1)
out-of-distribution generalization
(1)
information loss
(1)
recurrent neural network
(1)
language model
(1)
length generalization
(1)
quantitative reasoning
(1)
pretrained language model
(1)
long-range dependency
(1)
reasoning task
(1)
scientific reasoning
(1)
second order optimization
(1)
data whitening
(1)
second moment matrix
(1)
problem solving
(1)
Papers
Block-Recurrent Transformers
NIPS 2022
Exploring Length Generalization in Large Language Models
NIPS 2022
Solving Quantitative Reasoning Problems with Language Models
NIPS 2022
Effect of scale on catastrophic forgetting in neural networks
ICLR 2022
Whitening and Second Order Optimization Both Make Information in the Dataset Unusable During Training, and Can Reduce or Prevent Generalization
ICML 2021
Anatomy of Catastrophic Forgetting: Hidden Representations and Task Semantics
ICLR 2021
Tradeoffs in Data Augmentation: An Empirical Study
ICLR 2021
When Do Curricula Work?
ICLR 2021
Asymptotics of Wide Networks from Feynman Diagrams
ICLR 2020