Nikhil Vyas
11 papers · 2018–2025 · 4 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+6 more ↓ Show less ↑
🏃 Academic Marathon (7) 🌉 Interdisciplinary Bridge 🧭 Keyword Pioneer 🌍 Conference Polyglot (4) 🐝 Cross-Pollinator (13)
🧭
Keyword Pioneer
🌍
Conference Polyglot
(4)
🏃
Academic Marathon
(7)
⚡
Prolific Year
(5)
💎
Century Club
(11)
❓
The Questioner
Conferences
ICLR (5)
ICML (3)
NIPS (2)
EMNLP (1)
Top co-authors
Research topics
Keywords
representation learning
(1)
image classification
(1)
compressed sensing
(1)
neural network optimization
(1)
adversarial attack
(1)
generative model
(1)
diffusion model
(1)
mixture of expert
(1)
sparse activation
(1)
copyright protection
(1)
model scaling
(1)
adversarial example
(1)
expert routing
(1)
infinite width limit
(1)
routing mechanism
(1)
near access-freeness
(1)
sampling protected content
(1)
transformer model
(1)
finite width analysis
(1)
sparse fourier transform
(1)
Papers
Deconstructing What Makes a Good Optimizer for Autoregressive Language Models
ICLR 2025
A New Perspective on Shampoo's Preconditioner
ICLR 2025
How Does Critical Batch Size Scale in Pre-training?
ICLR 2025
SOAP: Improving and Stabilizing Shampoo using Adam for Language Modeling
ICLR 2025
Mixture of Parrots: Experts improve memorization more than reasoning
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
NIPS 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
NIPS 2018