Dan Roy
8 papers · 2011–2023 · 1 venue · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓
🏃
Academic Marathon · Half
🌉
Interdisciplinary Bridge · Bronze
🗺️
Taxonomy Completionist
(23)
🐣
Hot Topic Early Bird
📈
Trend Setter
Venues
Top co-authors
Research topics
Machine Learning
(8)
Optimization & Theory
(8)
Mathematics & Optimization
(3)
Artificial Intelligence
(2)
Deep Learning
(2)
Architectures
(2)
Learning Types
(2)
Mathematics
(2)
Knowledge & Reasoning
(1)
Natural Language Processing
(1)
Application Areas
(1)
Applications
(1)
Bayesian & Probabilistic · Artificial Intelligence
(1)
Bayesian & Probabilistic · Machine Learning
(1)
Core AI
(1)
Keywords
infinite-width limit
(2)
neural network initialization
(2)
stochastic differential equation
(2)
regret bound
(2)
probabilistic modeling
(1)
bayesian inference
(1)
neural network pruning
(1)
attention mechanism
(1)
online learning
(1)
topic modeling
(1)
computational complexity
(1)
minimax optimization
(1)
neural network theory
(1)
bayesian model
(1)
vc dimension
(1)
probabilistic inference
(1)
gaussian process
(1)
covariance matrix
(1)
map estimation
(1)
regret minimization
(1)
Papers
The Shaped Transformer: Attention Models in the Infinite Depth-and-Width Limit
NeurIPS 2023
Adaptively Exploiting d-Separators with Causal Bandits
NeurIPS 2022
Pruning’s Effect on Generalization Through the Lens of Training and Regularization
NeurIPS 2022
The Neural Covariance SDE: Shaped Infinite Depth-and-Width Networks at Initialization
NeurIPS 2022
Minimax Optimal Quantile and Semi-Adversarial Regret via Root-Logarithmic Regularizers
NeurIPS 2021
The future is log-Gaussian: ResNets and their infinite-depth-and-width limit at initialization
NeurIPS 2021
Towards a Unified Information-Theoretic Framework for Generalization
NeurIPS 2021
Complexity of Inference in Latent Dirichlet Allocation
NeurIPS 2011