Fanghui Liu
25 papers · 2020–2025 · 7 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 ↓+9 more ↓ Show less ↑
πΊοΈ Taxonomy Completionist (32) π Conference Polyglot Β· Silver π Academic Marathon Β· Sprint π Interdisciplinary Bridge Β· Bronze π£ Hot Topic Early Bird
π€
Dynamic Duo
(17)
π
Keyword Champion
(2)
π
Grand Slam
π¬
Deep Specialist
(14)
β‘
Prolific Year Β· Productive
π₯
Unstoppable Β· Bronze
β
The Questioner
π
Century Club Β· Bronze
ποΈ
Keyword Collector
(99)
Top co-authors
Research topics
Machine Learning
(23)
Optimization & Theory Β· Machine Learning
(19)
Deep Learning
(13)
Architectures
(7)
Core Methods
(7)
Optimization & Theory Β· Deep Learning
(6)
Learning Types
(5)
Artificial Intelligence
(3)
Mathematics & Optimization
(3)
Application Areas
(3)
Core AI
(2)
Optimization
(2)
Techniques
(2)
Natural Language Processing
(1)
Reinforcement Learning
(1)
Keywords
generalization bound
(6)
neural tangent kernel
(6)
neural network
(5)
kernel ridge regression
(4)
kernel methods
(4)
function approximation
(3)
reproducing kernel hilbert space
(3)
neural architecture search
(2)
barron space
(2)
double descent
(2)
kl divergence
(2)
bias-variance decomposition
(2)
deep neural network
(2)
random feature
(2)
gradient descent
(2)
adversarial training
(2)
regret bound
(2)
neural network robustness
(2)
sample complexity
(1)
feature learning
(1)
Papers
How Gradient descent balances features: A dynamical analysis for two-layer neural networks
ICLR 2025
LoRA-One: One-Step Full Gradient Could Suffice for Fine-Tuning Large Language Models, Provably and Efficiently
ICML 2025
Efficient local linearity regularization to overcome catastrophic overfitting
ICLR 2024
Generalization of Scaled Deep ResNets in the Mean-Field Regime
ICLR 2024
High-Dimensional Kernel Methods under Covariate Shift: Data-Dependent Implicit Regularization
ICML 2024
Learning Scalable Model Soup on a Single GPU: An Efficient Subspace Training Strategy
ECCV 2024
Learning with Norm Constrained, Over-parameterized, Two-layer Neural Networks
JMLR 2024
Revisiting Character-level Adversarial Attacks for Language Models
ICML 2024
Robust NAS under adversarial training: benchmark, theory, and beyond
ICLR 2024
The Role of Over-Parameterization in Machine Learning β the Good, the Bad, the Ugly
AAAI 2024
Benign Overfitting in Deep Neural Networks under Lazy Training
ICML 2023
Initialization Matters: Privacy-Utility Analysis of Overparameterized Neural Networks
NeurIPS 2023
On the Convergence of Encoder-only Shallow Transformers
NeurIPS 2023
What can online reinforcement learning with function approximation benefit from general coverage conditions?
ICML 2023
Extrapolation and Spectral Bias of Neural Nets with Hadamard Product: a Polynomial Net Study
NeurIPS 2022
Generalization Properties of NAS under Activation and Skip Connection Search
NeurIPS 2022
On the Double Descent of Random Features Models Trained with SGD
NeurIPS 2022
Robustness in deep learning: The good (width), the bad (depth), and the ugly (initialization)
NeurIPS 2022
Sound and Complete Verification of Polynomial Networks
NeurIPS 2022
Understanding Deep Neural Function Approximation in Reinforcement Learning via Ξ΅-Greedy Exploration
NeurIPS 2022
Fast Learning in Reproducing Kernel Krein Spaces via Signed Measures
AISTATS 2021
Generalization Properties of hyper-RKHS and its Applications
JMLR 2021
Kernel regression in high dimensions: Refined analysis beyond double descent
AISTATS 2021
Learning Data-adaptive Non-parametric Kernels
JMLR 2020
Random Fourier Features via Fast Surrogate Leverage Weighted Sampling
AAAI 2020