Nikhil Ghosh
9 papers · 2019–2025 · 5 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+2 more ↓ Show less ↑
π Cross-Pollinator (13) π Conference Polyglot (5) π Academic Marathon (6) π£ Hot Topic Early Bird π Interdisciplinary Bridge
πΊοΈ
Taxonomy Completionist
(15)
π
Triple Crown
Conferences
ICLR (3)
NIPS (3)
EMNLP (1)
ICML (1)
JMLR (1)
Top co-authors
Keywords
neural network optimization
(3)
transformer model
(2)
feature learning
(1)
sparse representation
(1)
computational efficiency
(1)
dimension reduction
(1)
euclidean embedding
(1)
bradley-terry-luce model
(1)
learning rate
(1)
stochastic gradient descent
(1)
sparse activation
(1)
model scaling
(1)
inference latency
(1)
expert routing
(1)
ordinal embedding
(1)
low-dimensional representation
(1)
routing mechanism
(1)
model capacity
(1)
low rank adaptation
(1)
mixture of expert
(1)
Papers
The Effect of SGD Batch Size on Autoencoder Learning: Sparsity, Sharpness, and Feature Learning
JMLR 2025
The Impact of Initialization on LoRA Finetuning Dynamics
NIPS 2024
More is Better: when Infinite Overparameterization is Optimal and Overfitting is Obligatory
ICLR 2024
LoRA+: Efficient Low Rank Adaptation of Large Models
ICML 2024
On the Benefits of Learning to Route in Mixture-of-Experts Models
EMNLP 2023
Alternating Updates for Efficient Transformers
NIPS 2023
Deconstructing Distributions: A Pointwise Framework of Learning
ICLR 2023
The Three Stages of Learning Dynamics in High-dimensional Kernel Methods
ICLR 2022
Landmark Ordinal Embedding
NIPS 2019