Mansheej Paul
8 papers · 2020–2025 · 3 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+3 more ↓ Show less ↑
🐝 Cross-Pollinator (11) 🌍 Conference Polyglot (3) 🏃 Academic Marathon (5) 🌉 Interdisciplinary Bridge 🗺️ Taxonomy Completionist (17)
🧭
Keyword Pioneer
🐣
Hot Topic Early Bird
❓
The Questioner
Conferences
NIPS (4)
ICLR (3)
ICML (1)
Top co-authors
Keywords
loss landscape
(2)
training dynamics
(2)
feature learning
(1)
in-context learning
(1)
kernel learning
(1)
linear regression
(1)
bayesian estimator
(1)
deep neural network
(1)
gradient norm
(1)
task diversity
(1)
sparse training
(1)
data pruning
(1)
loss surface
(1)
network initialization
(1)
iterative magnitude pruning
(1)
error norm
(1)
lottery ticket hypothesis
(1)
example importance
(1)
neural tangent kernel
(1)
Papers
$\mathrmμ$nit Scaling: Simple and Scalable FP8 LLM Training
ICML 2025
Scaling Laws for Precision
ICLR 2025
Perplexed by Perplexity: Perplexity-Based Data Pruning With Small Reference Models
ICLR 2025
Unmasking the Lottery Ticket Hypothesis: What's Encoded in a Winning Ticket's Mask?
ICLR 2023
Pretraining task diversity and the emergence of non-Bayesian in-context learning for regression
NIPS 2023
Lottery Tickets on a Data Diet: Finding Initializations with Sparse Trainable Networks
NIPS 2022
Deep Learning on a Data Diet: Finding Important Examples Early in Training
NIPS 2021
Deep learning versus kernel learning: an empirical study of loss landscape geometry and the time evolution of the Neural Tangent Kernel
NIPS 2020