Atish Agarwala
10 papers · 2021–2025 · 3 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+4 more ↓ Show less ↑
🐝 Cross-Pollinator (8) 🌉 Interdisciplinary Bridge 🌍 Conference Polyglot (3) 🧭 Keyword Pioneer 🐣 Hot Topic Early Bird
🏆
Keyword Champion
(2)
🔥
Unstoppable
(5)
💎
Century Club
(10)
❓
The Questioner
Conferences
ICML (6)
ICLR (2)
NIPS (2)
Top co-authors
Keywords
loss hessian
(2)
hessian eigenvalue
(2)
edge of stability
(2)
adaptive learning rate
(1)
learning rate
(1)
eigenvalue regularization
(1)
training stability
(1)
sharpness aware minimization
(1)
orthogonal matrix
(1)
gradient penalty
(1)
progressive sharpening
(1)
second-order regression model
(1)
deep equilibrium network
(1)
weight scale
(1)
gauss-newton matrix
(1)
second-order regression
(1)
optimization dynamics
(1)
learning rate tuning
(1)
sharpness regularization
(1)
neural tangent kernel
(1)
Papers
Avoiding spurious sharpness minimization broadens applicability of SAM
ICML 2025
Scaling Collapse Reveals Universal Dynamics in Compute-Optimally Trained Neural Networks
ICML 2025
Exact risk curves of signSGD in High-Dimensions: quantifying preconditioning and noise-compression effects
ICML 2025
To Clip or not to Clip: the Dynamics of SGD with Gradient Clipping in High-Dimensions
ICLR 2025
Stepping on the Edge: Curvature Aware Learning Rate Tuners
NIPS 2024
Neglected Hessian component explains mysteries in sharpness regularization
NIPS 2024
Second-order regression models exhibit progressive sharpening to the edge of stability
ICML 2023
SAM operates far from home: eigenvalue regularization as a dynamical phenomenon
ICML 2023
Deep equilibrium networks are sensitive to initialization statistics
ICML 2022
One Network Fits All? Modular versus Monolithic Task Formulations in Neural Networks
ICLR 2021