Sidak Pal Singh
15 papers · 2020–2025 · 5 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+7 more ↓ Show less ↑
๐บ๏ธ Taxonomy Completionist (18) ๐ Interdisciplinary Bridge ๐ Academic Marathon (5) ๐ Conference Polyglot (5) ๐งญ Keyword Pioneer
๐
Conference Polyglot
(5)
๐
Keyword Champion
(2)
๐
Grand Slam
๐ฅ
Unstoppable
(6)
โก
Prolific Year
(5)
๐
Century Club
(15)
โ
The Questioner
Conferences
ICLR (6)
NIPS (5)
ICML (2)
AAAI (1)
AISTATS (1)
Top co-authors
Keywords
neural network optimization
(3)
deep linear network
(2)
hessian matrix
(2)
model compression
(2)
model architecture
(2)
optimal transport
(2)
second-order optimization
(2)
knowledge transfer
(1)
knowledge distillation
(1)
network pruning
(1)
loss landscape
(1)
model fusion
(1)
hessian approximation
(1)
signal propagation
(1)
condition number
(1)
convolutional neural network
(1)
rank deficiency
(1)
feed-forward network
(1)
gradient vanishing
(1)
relu network
(1)
Papers
What Does It Mean to Be a Transformer? Insights from a Theoretical Hessian Analysis
ICLR 2025
Avoiding spurious sharpness minimization broadens applicability of SAM
ICML 2025
The Directionality of Optimization Trajectories in Neural Networks
ICLR 2025
Rethinking Attention: Exploring Shallow Feed-Forward Neural Networks as an Alternative to Attention Layers in Transformers (Student Abstract)
AAAI 2024
Some Fundamental Aspects about Lipschitz Continuity of Neural Networks
ICLR 2024
Towards Meta-Pruning via Optimal Transport
ICLR 2024
Theoretical Characterisation of the Gauss Newton Conditioning in Neural Networks
NIPS 2024
Transformer Fusion with Optimal Transport
ICLR 2024
The Hessian perspective into the Nature of Convolutional Neural Networks
ICML 2023
Phenomenology of Double Descent in Finite-Width Neural Networks
ICLR 2022
Signal Propagation in Transformers: Theoretical Perspectives and the Role of Rank Collapse
NIPS 2022
Analytic Insights into Structure and Rank of Neural Network Hessian Maps
NIPS 2021
WoodFisher: Efficient Second-Order Approximation for Neural Network Compression
NIPS 2020
Context Moverโs Distance & Barycenters: Optimal Transport of Contexts for Building Representations
AISTATS 2020
Model Fusion via Optimal Transport
NIPS 2020