Tin Sum Cheng

11 papers A* 3Journal 8
YearRankTypeTitle / Venue / Authors
2026 J jnl
Optimizer choice matters for the emergence of Neural Collapse.
CoRR
Jim Zhao, Tin Sum Cheng, Wojciech Masarczyk, Aurélien Lucchi
2025 J jnl
Beyond Universal Approximation Theorems: Algorithmic Uniform Approximation by Neural Networks Trained with Noisy Data.
CoRR
Anastasis Kratsios, Tin Sum Cheng, Daniel Roy
2025 J jnl
Sharp Generalization Bounds for Foundation Models with Asymmetric Randomized Low-Rank Adapters.
CoRR
Anastasis Kratsios, Tin Sum Cheng, Aurélien Lucchi, Haitz Sáez de Ocáriz Borde
2025 J jnl
Spectral Analysis of Molecular Kernels: When Richer Features Do Not Guarantee Better Generalization.
CoRR
Asma Jamali, Tin Sum Cheng, Rodrigo A. Vargas-Hernández
2025 J jnl
Unpacking Softmax: How Temperature Drives Representation Collapse, Compression, and Generalization.
CoRR
Wojciech Masarczyk, Mateusz Ostaszewski, Tin Sum Cheng, Tomasz Trzcinski, Aurélien Lucchi, Razvan Pascanu
2024 A* conf
A Comprehensive Analysis on the Learning Curve in Kernel Ridge Regression.
NeurIPS
Tin Sum Cheng, Aurélien Lucchi, Anastasis Kratsios, David Belius
2024 J jnl
A Comprehensive Analysis on the Learning Curve in Kernel Ridge Regression.
CoRR
Tin Sum Cheng, Aurélien Lucchi, Anastasis Kratsios, David Belius
2024 A* conf
Characterizing Overfitting in Kernel Ridgeless Regression Through the Eigenspectrum.
ICML
Tin Sum Cheng, Aurélien Lucchi, Anastasis Kratsios, David Belius
2024 J jnl
Characterizing Overfitting in Kernel Ridgeless Regression Through the Eigenspectrum.
CoRR
Tin Sum Cheng, Aurélien Lucchi, Anastasis Kratsios, David Belius
2023 A* conf
A Theoretical Analysis of the Test Error of Finite-Rank Kernel Ridge Regression.
NeurIPS
Tin Sum Cheng, Aurélien Lucchi, Anastasis Kratsios, Ivan Dokmanic, David Belius
2023 J jnl
A Theoretical Analysis of the Test Error of Finite-Rank Kernel Ridge Regression.
CoRR
Tin Sum Cheng, Aurélien Lucchi, Ivan Dokmanic, Anastasis Kratsios, David Belius