Nikhil Ghosh
9 papers · 2019–2025 · 5 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 ↓
🌍
Conference Polyglot · Silver
🏃
Academic Marathon · Sprint
🗺️
Taxonomy Completionist
(19)
🐣
Hot Topic Early Bird
👑
Triple Crown
Venues
Top co-authors
Research topics
Keywords
neural network optimization
(3)
learning rate
(2)
low-rank adaptation
(2)
transformer model
(2)
stochastic gradient descent
(2)
dimension reduction
(1)
generalization error
(1)
kernel regression
(1)
euclidean embedding
(1)
bradley-terry-luce model
(1)
training dynamics
(1)
mixture of expert
(1)
sparse activation
(1)
data distribution
(1)
model scaling
(1)
inference latency
(1)
expert routing
(1)
random feature regression
(1)
model accuracy
(1)
ordinal embedding
(1)
Papers
The Effect of SGD Batch Size on Autoencoder Learning: Sparsity, Sharpness, and Feature Learning
JMLR 2025
LoRA+: Efficient Low Rank Adaptation of Large Models
ICML 2024
More is Better: when Infinite Overparameterization is Optimal and Overfitting is Obligatory
ICLR 2024
The Impact of Initialization on LoRA Finetuning Dynamics
NeurIPS 2024
Alternating Updates for Efficient Transformers
NeurIPS 2023
Deconstructing Distributions: A Pointwise Framework of Learning
ICLR 2023
On the Benefits of Learning to Route in Mixture-of-Experts Models
EMNLP 2023
The Three Stages of Learning Dynamics in High-dimensional Kernel Methods
ICLR 2022
Landmark Ordinal Embedding
NeurIPS 2019