Nikhil Vyas
11 papers · 2018–2025 · 4 venues · 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 ↓+4 more ↓ Show less ↑
π Interdisciplinary Bridge Β· Bronze πΊοΈ Taxonomy Completionist (28) π Conference Polyglot Β· Bronze π£ Hot Topic Early Bird π Academic Marathon Β· Sprint
β‘
Prolific Year Β· Productive
β
The Questioner
π
Century Club Β· Bronze
ποΈ
Keyword Collector
(51)
Venues
Top co-authors
Research topics
Machine Learning
(9)
Deep Learning
(7)
Optimization & Theory Β· Machine Learning
(7)
Core Methods
(5)
Artificial Intelligence
(3)
Core AI
(3)
Mathematics & Optimization
(2)
Natural Language Processing
(2)
Architectures
(2)
Learning Types
(2)
Models
(2)
Optimization
(2)
Optimization & Theory Β· Deep Learning
(2)
Techniques
(2)
Security & Privacy
(1)
Keywords
language model
(2)
mixture of expert
(2)
expert routing
(2)
neural network optimization
(2)
epistemic uncertainty
(1)
diffusion model
(1)
compressed sensing
(1)
adversarial attack
(1)
embedding
(1)
generative model
(1)
learning rate
(1)
optimization algorithm
(1)
second-order optimization
(1)
autoregressive model
(1)
adam optimizer
(1)
hessian approximation
(1)
implicit bia
(1)
scaling law
(1)
online learning
(1)
image classification
(1)
Papers
A New Perspective on Shampoo's Preconditioner
ICLR 2025
Deconstructing What Makes a Good Optimizer for Autoregressive Language Models
ICLR 2025
How Does Critical Batch Size Scale in Pre-training?
ICLR 2025
Mixture of Parrots: Experts improve memorization more than reasoning
ICLR 2025
SOAP: Improving and Stabilizing Shampoo using Adam for Language Modeling
ICLR 2025
Beyond Implicit Bias: The Insignificance of SGD Noise in Online Learning
ICML 2024
Distinguishing the Knowable from the Unknowable with Language Models
ICML 2024
Feature-Learning Networks Are Consistent Across Widths At Realistic Scales
NeurIPS 2023
On Provable Copyright Protection for Generative Models
ICML 2023
On the Benefits of Learning to Route in Mixture-of-Experts Models
EMNLP 2023
Thwarting Adversarial Examples: An L_0-Robust Sparse Fourier Transform
NeurIPS 2018