Sidak Pal Singh

42 papers A* 14A 2Journal 23Unranked 3
YearRankTypeTitle / Venue / Authors
2025 J jnl
Accelerating Neural Network Training Along Sharp and Flat Directions.
CoRR
Daniyar Zakarin, Sidak Pal Singh
2025 A* conf
Avoiding spurious sharpness minimization broadens applicability of SAM.
ICML
Sidak Pal Singh, Hossein Mobahi, Atish Agarwala, Yann N. Dauphin
2025 J jnl
Avoiding spurious sharpness minimization broadens applicability of SAM.
CoRR
Sidak Pal Singh, Hossein Mobahi, Atish Agarwala, Yann N. Dauphin
2025 J jnl
Generalized Linear Mode Connectivity for Transformers.
CoRR
Alexander Theus, Alessandro Cabodi, Sotiris Anagnostidis, Antonio Orvieto, Sidak Pal Singh, Valentina Boeva
2025 J jnl
Model Fusion via Neuron Interpolation.
CoRR
Phoomraphee Luenam, Andreas Spanopoulos, Amit Sant, Thomas Hofmann, Sotiris Anagnostidis, Sidak Pal Singh
2025 A* conf
The Directionality of Optimization Trajectories in Neural Networks.
ICLR
Sidak Pal Singh, Bobby He, Thomas Hofmann, Bernhard Schölkopf
2025 A* conf
What Does It Mean to Be a Transformer? Insights from a Theoretical Hessian Analysis.
ICLR
Weronika Ormaniec, Felix Dangel, Sidak Pal Singh
2024 J jnl
Hallmarks of Optimization Trajectories in Neural Networks and LLMs: The Lengths, Bends, and Dead Ends.
CoRR
Sidak Pal Singh, Bobby He, Thomas Hofmann, Bernhard Schölkopf
2024 J jnl
Landscaping Linear Mode Connectivity.
CoRR
Sidak Pal Singh, Linara Adilova, Michael Kamp, Asja Fischer, Bernhard Schölkopf, Thomas Hofmann
2024 — conf
Local vs Global continual learning.
CoLLAs
Giulia Lanzillotta, Sidak Pal Singh, Benjamin F. Grewe, Thomas Hofmann
2024 J jnl
Local vs Global continual learning.
CoRR
Giulia Lanzillotta, Sidak Pal Singh, Benjamin F. Grewe, Thomas Hofmann
2024 A* conf
Rethinking Attention: Exploring Shallow Feed-Forward Neural Networks as an Alternative to Attention Layers in Transformers (Student Abstract).
AAAI
Danilo Dordevic, Vukasin Bozic, Joseph Thommes, Daniele Coppola, Sidak Pal Singh
2024 A* conf
Some Fundamental Aspects about Lipschitz Continuity of Neural Networks.
ICLR
Grigory Khromov, Sidak Pal Singh
2024 A* conf
Theoretical Characterisation of the Gauss Newton Conditioning in Neural Networks.
NeurIPS
Jim Zhao, Sidak Pal Singh, Aurélien Lucchi
2024 J jnl
Theoretical characterisation of the Gauss-Newton conditioning in Neural Networks.
CoRR
Jim Zhao, Sidak Pal Singh, Aurélien Lucchi
2024 A* conf
Towards Meta-Pruning via Optimal Transport.
ICLR
Alexander Theus, Olin Geimer, Friedrich Wicke, Thomas Hofmann, Sotiris Anagnostidis, Sidak Pal Singh
2024 J jnl
Towards Meta-Pruning via Optimal Transport.
CoRR
Alexander Theus, Olin Geimer, Friedrich Wicke, Thomas Hofmann, Sotiris Anagnostidis, Sidak Pal Singh
2024 A* conf
Transformer Fusion with Optimal Transport.
ICLR
Moritz Imfeld, Jacopo Graldi, Marco Giordano, Thomas Hofmann, Sotiris Anagnostidis, Sidak Pal Singh
2024 J jnl
What Does It Mean to Be a Transformer? Insights from a Theoretical Hessian Analysis.
CoRR
Weronika Ormaniec, Felix Dangel, Sidak Pal Singh
2023 J jnl
On the curvature of the loss landscape.
CoRR
Alison Pouplin, Hrittik Roy, Sidak Pal Singh, Georgios Arvanitidis
2023 J jnl
Rethinking Attention: Exploring Shallow Feed-Forward Neural Networks as an Alternative to Attention Layers in Transformers.
CoRR
Vukasin Bozic, Danilo Dordevic, Daniele Coppola, Joseph Thommes, Sidak Pal Singh
2023 J jnl
Some Fundamental Aspects about Lipschitz Continuity of Neural Network Functions.
CoRR
Grigory Khromov, Sidak Pal Singh
2023 A* conf
The Hessian perspective into the Nature of Convolutional Neural Networks.
ICML
Sidak Pal Singh, Thomas Hofmann, Bernhard Schölkopf
2023 J jnl
The Hessian perspective into the Nature of Convolutional Neural Networks.
CoRR
Sidak Pal Singh, Thomas Hofmann, Bernhard Schölkopf
2023 J jnl
Towards guarantees for parameter isolation in continual learning.
CoRR
Giulia Lanzillotta, Sidak Pal Singh, Benjamin F. Grewe, Thomas Hofmann
2023 J jnl
Transformer Fusion with Optimal Transport.
CoRR
Moritz Imfeld, Jacopo Graldi, Marco Giordano, Thomas Hofmann, Sotiris Anagnostidis, Sidak Pal Singh
2022 A* conf
Phenomenology of Double Descent in Finite-Width Neural Networks.
ICLR
Sidak Pal Singh, Aurélien Lucchi, Thomas Hofmann, Bernhard Schölkopf
2022 J jnl
Phenomenology of Double Descent in Finite-Width Neural Networks.
CoRR
Sidak Pal Singh, Aurélien Lucchi, Thomas Hofmann, Bernhard Schölkopf
2022 A* conf
Signal Propagation in Transformers: Theoretical Perspectives and the Role of Rank Collapse.
NeurIPS
Lorenzo Noci, Sotiris Anagnostidis, Luca Biggio, Antonio Orvieto, Sidak Pal Singh, Aurélien Lucchi
2022 J jnl
Signal Propagation in Transformers: Theoretical Perspectives and the Role of Rank Collapse.
CoRR
Lorenzo Noci, Sotiris Anagnostidis, Luca Biggio, Antonio Orvieto, Sidak Pal Singh, Aurélien Lucchi
2021 A* conf
Analytic Insights into Structure and Rank of Neural Network Hessian Maps.
NeurIPS
Sidak Pal Singh, Gregor Bachmann, Thomas Hofmann
2021 J jnl
Analytic Insights into Structure and Rank of Neural Network Hessian Maps.
CoRR
Sidak Pal Singh, Gregor Bachmann, Thomas Hofmann
2020 A conf
Context Mover's Distance & Barycenters: Optimal Transport of Contexts for Building Representations.
AISTATS
Sidak Pal Singh, Andreas Hug, Aymeric Dieuleveut, Martin Jaggi
2020 A* conf
Model Fusion via Optimal Transport.
NeurIPS
Sidak Pal Singh, Martin Jaggi
2020 A* conf
WoodFisher: Efficient Second-Order Approximation for Neural Network Compression.
NeurIPS
Sidak Pal Singh, Dan Alistarh
2020 J jnl
WoodFisher: Efficient second-order approximations for model compression.
CoRR
Sidak Pal Singh, Dan Alistarh
2019 — conf
Context Mover's Distance & Barycenters: Optimal transport of contexts for building representations.
DGS@ICLR
Sidak Pal Singh, Andreas Hug, Aymeric Dieuleveut, Martin Jaggi
2019 J jnl
GLOSS: Generative Latent Optimization of Sentence Representations.
CoRR
Sidak Pal Singh, Angela Fan, Michael Auli
2019 J jnl
Model Fusion via Optimal Transport.
CoRR
Sidak Pal Singh, Martin Jaggi
2018 J jnl
Wasserstein is all you need.
CoRR
Sidak Pal Singh, Andreas Hug, Aymeric Dieuleveut, Martin Jaggi
2017 A conf
RaaS and Hierarchical Aggregation Revisited.
ICWS
Rohit Ranchal, Sidak Pal Singh, Pelin Angin, Ajay Mohindra, Hui Lei, Bharat K. Bhargava
2016 — conf
SL - FII: Syntactic and Lexical Constraints with Frequency based Iterative Improvement for Disease Mention Recognition in News Headlines.
BAI@IJCAI
Sidak Pal Singh, Sopan Khosla, Sajal Rustagi, Manisha Patel, Dhaval Patel